row_id string | section string | section_slug string | resource_type string | marker string | title string | url string | url_kind string | domain string | annotation string | description string | key_contribution string | novelty string | impact string | signal string | signal_strength string | source_readme string | source_line int64 | source_url string | date_added string | collection string | collection_slug string | user_goal string | lifecycle_stages string | audience string | loop_layer string | scope_fit string | evidence_class string | evidence_tier string | source_status string | canonical_url string | source_title string | source_description string | authors string | publication_date string | publication_year string | publication_venue string | publisher string | doi string | publication_note string | primary_category string | metadata_source string | github_repo string | github_stars string | github_forks string | github_license string | github_created_at string | github_updated_at string | arxiv_id string | audited_at timestamp[ms] |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
ale-0701 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Restate | https://restate.dev/ | external | restate.dev | Durable execution runtime for building resilient, stateful agents and workflows that survive failures mid-loop. | Durable execution runtime for building resilient, stateful agents and workflows that survive failures mid-loop. | Durable execution runtime for building resilient, stateful agents and workflows that survive failures mid-loop. | Durable execution and replay are treated as first-class loop infrastructure. Durable execution runtime for building resilient, stateful agents and workflows that survive failures mid-loop. | Use Restate to choose an implementation surface for repeatable agent work. | Working implementation, framework, runtime, or repository; signal comes from usable code and ecosystem adoption. | high | README.md | 1,398 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1398 | Build | build | Choose runtimes, tools, and delegation surfaces. | state | builder | workflow | enabling | implementation | A | ok | https://restate.dev/ | Restate - Build innately resilient distributed apps | Restate is a lightweight runtime that lets developers build innately resilient distributed apps without the complexity tax. | Restate | html-meta | 2026-09-08T16:27:48 | |||||||||||||||
ale-0702 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | DBOS | https://www.dbos.dev/ | external | www.dbos.dev | Lightweight PostgreSQL-backed durable execution library for crash-proof agent workflows, queues, and scheduled triggers. | Lightweight PostgreSQL-backed durable execution library for crash-proof agent workflows, queues, and scheduled triggers. | Lightweight PostgreSQL-backed durable execution library for crash-proof agent workflows, queues, and scheduled triggers. | Durable execution and replay are treated as first-class loop infrastructure. Lightweight PostgreSQL-backed durable execution library for crash-proof agent workflows, queues, and scheduled triggers. | Use DBOS to choose an implementation surface for repeatable agent work. | Working implementation, framework, runtime, or repository; signal comes from usable code and ecosystem adoption. | high | README.md | 1,399 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1399 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;intake | builder | workflow | enabling | implementation | A | ok | https://www.dbos.dev/ | DBOS | Durable Workflow Orchestration | DBOS is an open source durable execution and workflow orchestration system that radically simplifies the development and operation of reliable, observable workflows. | dbos.dev | domain-fallback | 2026-09-08T16:27:48 | |||||||||||||||
ale-0703 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Composio Agent Orchestrator | https://github.com/ComposioHQ/agent-orchestrator | external | github.com | Orchestrates parallel coding agents in isolated worktrees that plan tasks, fix CI failures, respond to reviews, and manage their own PR lifecycle. | Orchestrates parallel coding agents in isolated worktrees that plan tasks, fix CI failures, respond to reviews, and manage their own PR lifecycle. | Orchestrates parallel coding agents in isolated worktrees that plan tasks, fix CI failures, respond to reviews, and manage their own PR lifecycle. | Workspace isolation is part of the loop design, not an afterthought. Orchestrates parallel coding agents in isolated worktrees that plan tasks, fix CI failures, respond to reviews, and manage their own PR lifecycle. | Use Composio Agent Orchestrator to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,400 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1400 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;delegation | builder | workflow | enabling | source-implementation | A | ok | https://github.com/Untrivial-ai/agent-orchestrator | GitHub - Untrivial-ai/agent-orchestrator: Agent IDE that enables you to manage fleets of coding agents. It comes with an agentic orchestrator that plans tasks, spawns agents, and autonomously handles CI fixes, merge conflicts, and code reviews. · GitHub | Agent IDE that enables you to manage fleets of coding agents. It comes with an agentic orchestrator that plans tasks, spawns agents, and autonomously handles CI fixes, merge conflicts, and code reviews. - Untrivial-ai/agent-orchestrator | ComposioHQ/agent-orchestrator | GitHub | html-meta | ComposioHQ/agent-orchestrator | 2026-09-08T16:27:48 | |||||||||||||
ale-0704 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Omnigent | https://github.com/omnigent-ai/omnigent | external | github.com | Databricks' open-source meta-harness and control plane that runs Claude Code, Codex, Cursor, and Pi under shared policies, with budget caps and human-approval gates enforced at the harness layer rather than in prompts. | Databricks' open-source meta-harness and control plane that runs Claude Code, Codex, Cursor, and Pi under shared policies, with budget caps and human-approval gates enforced at the harness layer rather than in prompts. | Databricks' open-source meta-harness and control plane that runs Claude Code, Codex, Cursor, and Pi under shared policies, with budget caps and human-approval gates enforced at the harness layer rather than in prompts. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Databricks' open-source meta-harness and control plane that runs Claude Code, Codex, Cursor, and Pi under shared policies, with budget caps and human-approval gates enforced at the harness layer rather than in prompts. | Use Omnigent to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,401 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1401 | Build | build | Choose runtimes, tools, and delegation surfaces. | budget;escalation | builder | workflow | enabling | source-implementation | A | ok | https://github.com/omnigent-ai/omnigent | GitHub - omnigent-ai/omnigent: Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device. · GitHub | Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device. - omnigent-ai/omnigent | omnigent-ai/omnigent | GitHub | html-meta | omnigent-ai/omnigent | 2026-09-08T16:27:48 | |||||||||||||
ale-0705 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | From Agent Loops to Structured Graphs: A Scheduler-Theoretic Framework for LLM Agent Execution | https://arxiv.org/abs/2604.11378 | external | arxiv.org | Replaces opaque agent loops with immutable plan-version DAGs and a planning-execution-recovery split, giving inspectable scheduling, deterministic recovery, escalation, and termination guarantees. | Replaces opaque agent loops with immutable plan-version DAGs and a planning-execution-recovery split, giving inspectable scheduling, deterministic recovery, escalation, and termination guarantees. | Replaces opaque agent loops with immutable plan-version DAGs and a planning-execution-recovery split, giving inspectable scheduling, deterministic recovery, escalation, and termination guarantees. | Control flow is represented as an inspectable graph rather than an opaque prompt loop. Replaces opaque agent loops with immutable plan-version DAGs and a planning-execution-recovery split, giving inspectable scheduling, deterministic recovery, escalation, and termination guarantees. | Use From Agent Loops to Structured Graphs: A Scheduler-Theoretic Framework for LLM Agent Execution to choose an implementation surface for repeatable agent work. | Research source arXiv:2604.11378; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,402 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1402 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;escalation;exit | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2604.11378 | [2604.11378] From Agent Loops to Structured Graphs:A Scheduler-Theoretic Framework for LLM Agent Execution | The dominant paradigm for building LLM based agents is the Agent Loop, an iterative cycle where a single language model decides what to do next by reading an ever growing context window. This paradigm has three structural weaknesses: implicit dependencies between steps, unbounded recovery loops, and mutable execution h... | Hu Wei | 2026-04-13 | 2026 | arXiv | arXiv | 51 pages, 4 figures | cs.AI | arxiv-api | 2604.11378 | 2026-09-08T16:27:48 | ||||||||
ale-0706 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Eve | https://github.com/vercel/eve | external | github.com | Vercel's TypeScript-native agent framework with durable execution, sandboxed compute, and OpenTelemetry tracing built in, so recurring agent work persists, replays, and is observable across runs by default. | Vercel's TypeScript-native agent framework with durable execution, sandboxed compute, and OpenTelemetry tracing built in, so recurring agent work persists, replays, and is observable across runs by default. | Vercel's TypeScript-native agent framework with durable execution, sandboxed compute, and OpenTelemetry tracing built in, so recurring agent work persists, replays, and is observable across runs by default. | Durable execution and replay are treated as first-class loop infrastructure. Vercel's TypeScript-native agent framework with durable execution, sandboxed compute, and OpenTelemetry tracing built in, so recurring agent work persists, replays, and is observable across runs by default. | Use Eve to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,403 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1403 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/vercel/eve | GitHub - vercel/eve: The Open Framework for Building Agents · GitHub | The Open Framework for Building Agents. Contribute to vercel/eve development by creating an account on GitHub. | vercel/eve | GitHub | html-meta | vercel/eve | 2026-09-08T16:27:48 | |||||||||||||
ale-0707 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework | https://arxiv.org/abs/2603.11445 | external | arxiv.org | Decomposes work into a dependency-aware DAG, runs domain agents in parallel, and uses an LLM verifier to drive adaptive replanning with configurable stop conditions, the verify-and-replan core of a reliable loop. | Decomposes work into a dependency-aware DAG, runs domain agents in parallel, and uses an LLM verifier to drive adaptive replanning with configurable stop conditions, the verify-and-replan core of a reliable loop. | Decomposes work into a dependency-aware DAG, runs domain agents in parallel, and uses an LLM verifier to drive adaptive replanning with configurable stop conditions, the verify-and-replan core of a reliable loop. | Control flow is represented as an inspectable graph rather than an opaque prompt loop. Decomposes work into a dependency-aware DAG, runs domain agents in parallel, and uses an LLM verifier to drive adaptive replanning with configurable stop conditions, the verify-and-replan core of a reliable loop. | Use Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework to choose an implementation surface for repeatable agent work. | Research source arXiv:2603.11445; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,404 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1404 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;verification;exit | researcher;evaluator | workflow | enabling | research-paper | A | ok | https://openreview.net/forum?id=WUmz4LUbvU | [2603.11445] Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution | We present Verified Multi-Agent Orchestration (VMAO), a framework that coordinates specialized LLM-based agents through a verification-driven iterative loop. Given a complex query, our system decomposes it into a directed acyclic graph (DAG) of sub-questions, executes them through domain-specific agents in parallel, ve... | Xing Zhang; Yanwei Cui; Guanghui Wang; Wei Qiu; Ziyuan Li; Fangwei Han; Yajing Huang; Hengzhi Qiu; Bing Zhu; Peiyang He | 2026 | 2026 | ICLR Workshop on Multi-Agent Learning: Generalization and Adaptation in Intelligence (MALGAI) | International Conference on Learning Representations | Published in ICLR Workshop on Multi-Agent Learning: Generalization and Adaptation in Intelligence (MALGAI); the linked arXiv record remains available for open access. | cs.AI | ICLR workshop OpenReview record | 2603.11445 | 2026-09-08T16:27:48 | ||||||||
ale-0708 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents | https://arxiv.org/abs/2603.22386 | external | arxiv.org | Organizes how agent workflows are fixed ahead of time or generated and revised per run, and which evaluation signals drive that choice, a map of the design space for recurring loops. | Organizes how agent workflows are fixed ahead of time or generated and revised per run, and which evaluation signals drive that choice, a map of the design space for recurring loops. | Organizes how agent workflows are fixed ahead of time or generated and revised per run, and which evaluation signals drive that choice, a map of the design space for recurring loops. | Control flow is represented as an inspectable graph rather than an opaque prompt loop. Organizes how agent workflows are fixed ahead of time or generated and revised per run, and which evaluation signals drive that choice, a map of the design space for recurring loops. | Use From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents to choose an implementation surface for repeatable agent work. | Research source arXiv:2603.22386; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,405 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1405 | Build | build | Choose runtimes, tools, and delegation surfaces. | verification | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2603.22386 | [2603.22386] From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents | Large language model (LLM)-based systems are becoming increasingly popular for solving tasks by constructing executable workflows that interleave LLM calls, information retrieval, tool use, code execution, memory updates, and verification. This survey reviews recent methods for designing and optimizing such workflows, ... | Ling Yue; Kushal Raj Bhandari; Ching-Yun Ko; Dhaval Patel; Shuxin Lin; Nianjun Zhou; Jianxi Gao; Pin-Yu Chen; Shaowu Pan | 2026-03-23 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2603.22386 | 2026-09-08T16:27:48 | |||||||||
ale-0709 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Agent-as-a-Router | https://github.com/LanceZPF/agent-as-a-router | external | github.com | Agentic model routing for coding agents reframed as a context-action-feedback loop (ACRouter: orchestrator, verifier, memory) that learns which LLM to route each task to from execution feedback rather than frozen priors, with the CodeRouterBench benchmark across 8 frontier models. | Agentic model routing for coding agents reframed as a context-action-feedback loop (ACRouter: orchestrator, verifier, memory) that learns which LLM to route each task to from execution feedback rather than frozen priors, with the CodeRouterBench benchmark across 8 frontier models. | Agentic model routing for coding agents reframed as a context-action-feedback loop (ACRouter: orchestrator, verifier, memory) that learns which LLM to route each task to from execution feedback rather than frozen priors, with the CodeRouterBench benchmark across 8 frontier models. | Verification is promoted from a final check to a loop-control signal. Agentic model routing for coding agents reframed as a context-action-feedback loop (ACRouter: orchestrator, verifier, memory) that learns which LLM to route each task to from execution feedback rather than frozen priors, with the CodeRouterBench benc... | Use Agent-as-a-Router to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,406 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1406 | Build | build | Choose runtimes, tools, and delegation surfaces. | context;delegation;verification | builder | workflow | enabling | source-implementation | A | ok | https://github.com/LanceZPF/agent-as-a-router | GitHub - LanceZPF/agent-as-a-router: The official implementations of Agent-as-a-Router: Agentic Model Routing for Coding Tasks. · GitHub | The official implementations of Agent-as-a-Router: Agentic Model Routing for Coding Tasks. - LanceZPF/agent-as-a-router | LanceZPF/agent-as-a-router | GitHub | html-meta | LanceZPF/agent-as-a-router | 2026-09-08T16:27:48 | |||||||||||||
ale-0710 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Amp: Custom Agents | https://ampcode.com/news/custom-agents | external | ampcode.com | Amp's plugin-defined custom agents that run as the main agent or as subagents, spawn parallel workers, join tool pipelines, and use thread actions to build background review threads that report results back to a parent thread. | Amp's plugin-defined custom agents that run as the main agent or as subagents, spawn parallel workers, join tool pipelines, and use thread actions to build background review threads that report results back to a parent thread. | Amp's plugin-defined custom agents that run as the main agent or as subagents, spawn parallel workers, join tool pipelines, and use thread actions to build background review threads that report results back to a parent thread. | The work separates roles across agents, verifiers, or orchestration layers. Amp's plugin-defined custom agents that run as the main agent or as subagents, spawn parallel workers, join tool pipelines, and use thread actions to build background review threads that report results back to a parent thread. | Use Amp: Custom Agents to choose an implementation surface for repeatable agent work. | Contextual source from ampcode.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,407 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1407 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;delegation | builder | workflow | enabling | practitioner-analysis | B | ok | https://ampcode.com/news/custom-agents | Custom Agents - Amp | Plugins can now create agents, run them once, and keep talking to their threads. | Amp | 2026-06-19 | 2026 | ampcode.com | html-meta | 2026-09-08T16:27:48 | ||||||||||||
ale-0711 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | AgentsMesh | https://github.com/AgentsMesh/AgentsMesh | external | github.com | Self-hosted control plane for running fleets of coding agents across your own machines, with scheduling, per-pod Git worktree isolation, Kanban work tracking, and merge-request integration. | Self-hosted control plane for running fleets of coding agents across your own machines, with scheduling, per-pod Git worktree isolation, Kanban work tracking, and merge-request integration. | Self-hosted control plane for running fleets of coding agents across your own machines, with scheduling, per-pod Git worktree isolation, Kanban work tracking, and merge-request integration. | Workspace isolation is part of the loop design, not an afterthought. Self-hosted control plane for running fleets of coding agents across your own machines, with scheduling, per-pod Git worktree isolation, Kanban work tracking, and merge-request integration. | Use AgentsMesh to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,408 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1408 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;workspace | builder | workflow | enabling | source-implementation | A | ok | https://github.com/AgentsMesh/AgentsMesh | GitHub - AgentsMesh/AgentsMesh: The AI Agent Workforce Platform. Run a hundred AI coding agents across your own machines — schedule, isolate, and steer them all from one console. · GitHub | The AI Agent Workforce Platform. Run a hundred AI coding agents across your own machines — schedule, isolate, and steer them all from one console. - AgentsMesh/AgentsMesh | AgentsMesh/AgentsMesh | GitHub | html-meta | AgentsMesh/AgentsMesh | 2026-09-08T16:27:48 | |||||||||||||
ale-0712 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Bernstein | https://github.com/sipyourdrink-ltd/bernstein | external | github.com | Deterministic Python orchestrator that runs parallel CLI coding agents in isolated Git worktrees, gates merges on tests, lint, and type checks, and records every scheduling decision in a tamper-evident audit log. | Deterministic Python orchestrator that runs parallel CLI coding agents in isolated Git worktrees, gates merges on tests, lint, and type checks, and records every scheduling decision in a tamper-evident audit log. | Deterministic Python orchestrator that runs parallel CLI coding agents in isolated Git worktrees, gates merges on tests, lint, and type checks, and records every scheduling decision in a tamper-evident audit log. | Workspace isolation is part of the loop design, not an afterthought. Deterministic Python orchestrator that runs parallel CLI coding agents in isolated Git worktrees, gates merges on tests, lint, and type checks, and records every scheduling decision in a tamper-evident audit log. | Use Bernstein to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,409 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1409 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;workspace;delegation;verification | builder | workflow | enabling | source-implementation | A | ok | https://github.com/sipyourdrink-ltd/bernstein | GitHub - sipyourdrink-ltd/bernstein: The open‑source AI Agents Governance & Orchestration framework: write the rules declaratively, Bernstein enforces them and produces the verifiable, replayable record. Free, Apache-2.0. https://bernstein.run · GitHub | The open‑source AI Agents Governance & Orchestration framework: write the rules declaratively, Bernstein enforces them and produces the verifiable, replayable record. Free, Apache-2.0. https://bernstein.run - sipyourdrink-ltd/bernstein | sipyourdrink-ltd/bernstein | GitHub | html-meta | sipyourdrink-ltd/bernstein | 2026-09-08T16:27:48 | |||||||||||||
ale-0713 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Aeon | https://github.com/aaronjmars/aeon | external | github.com | Autonomous agent framework that runs Claude Code unattended on GitHub Actions, dispatching skills on cron or reactive triggers with per-run quality scoring, persistent memory, and self-healing skill repair. | Autonomous agent framework that runs Claude Code unattended on GitHub Actions, dispatching skills on cron or reactive triggers with per-run quality scoring, persistent memory, and self-healing skill repair. | Autonomous agent framework that runs Claude Code unattended on GitHub Actions, dispatching skills on cron or reactive triggers with per-run quality scoring, persistent memory, and self-healing skill repair. | Persistent memory is treated as an external runtime artifact. Autonomous agent framework that runs Claude Code unattended on GitHub Actions, dispatching skills on cron or reactive triggers with per-run quality scoring, persistent memory, and self-healing skill repair. | Use Aeon to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,410 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1410 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;context;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/aeonfun/aeon | GitHub - aeonfun/aeon: The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever. · GitHub | The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever. - aeonfun/aeon | aaronjmars/aeon | GitHub | html-meta | aaronjmars/aeon | 2026-09-08T16:27:48 | |||||||||||||
ale-0714 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | h5i | https://github.com/h5i-dev/h5i | external | github.com | Gives each coding agent an isolated sandboxed Git worktree, dispatches one task to a team that peer-reviews each other's candidates, then replays and tests each candidate with a neutral verifier before merging the winner. | Gives each coding agent an isolated sandboxed Git worktree, dispatches one task to a team that peer-reviews each other's candidates, then replays and tests each candidate with a neutral verifier before merging the winner. | Gives each coding agent an isolated sandboxed Git worktree, dispatches one task to a team that peer-reviews each other's candidates, then replays and tests each candidate with a neutral verifier before merging the winner. | Workspace isolation is part of the loop design, not an afterthought. Gives each coding agent an isolated sandboxed Git worktree, dispatches one task to a team that peer-reviews each other's candidates, then replays and tests each candidate with a neutral verifier before merging the winner. | Use h5i to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,411 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1411 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;verification | builder | workflow | enabling | source-implementation | A | ok | https://github.com/h5i-dev/h5i | GitHub - h5i-dev/h5i: Fast, security-first headless browser for AI agents. Built for automation, red teaming, web testing, and scraping, with direct HTTP traffic control and auditable sessions. Pure Rust, no Chromium or V8. · GitHub | Fast, security-first headless browser for AI agents. Built for automation, red teaming, web testing, and scraping, with direct HTTP traffic control and auditable sessions. Pure Rust, no Chromium or V8. - h5i-dev/h5i | h5i-dev/h5i | GitHub | html-meta | h5i-dev/h5i | 2026-09-08T16:27:48 | |||||||||||||
ale-0715 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery | https://arxiv.org/abs/2607.02807 | external | arxiv.org | A shepherd agent with global context steers a population of search agents that each work in their own Git branch with local context, countering the context accumulation of solo long-running agents and matching or beating baselines on 13 of 15 open-ended discovery tasks. | A shepherd agent with global context steers a population of search agents that each work in their own Git branch with local context, countering the context accumulation of solo long-running agents and matching or beating baselines on 13 of 15 open-ended discovery tasks. | A shepherd agent with global context steers a population of search agents that each work in their own Git branch with local context, countering the context accumulation of solo long-running agents and matching or beating baselines on 13 of 15 open-ended discovery tasks. | Context is managed as durable loop state rather than a single prompt payload. A shepherd agent with global context steers a population of search agents that each work in their own Git branch with local context, countering the context accumulation of solo long-running agents and matching or beating baselines on 13 of 15... | Use SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.02807; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,412 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1412 | Build | build | Choose runtimes, tools, and delegation surfaces. | intake;context | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.02807 | [2607.02807] SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery | Long-running coding agents such as autoresearch can persistently discover optimizations for open-ended problems. However, they tend to converge onto a single high-level approach, then proceed with low-level edits while missing other superior approaches to the problem. We hypothesize two harness-level design choices con... | Yuvraj Virk; Zack Edds; Chunqiu Steven Xia; Lingming Zhang | 2026-07-02 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2607.02807 | 2026-09-08T16:27:48 | |||||||||
ale-0716 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Scaling Long-Running Autonomous Coding | https://cursor.com/blog/scaling-agents | external | cursor.com | Cursor traces the coordination designs behind week-long autonomous coding runs, from flat agents with locking to optimistic concurrency to a planner/worker/judge hierarchy, letting hundreds of concurrent workers push to one branch on projects exceeding a million lines. | Cursor traces the coordination designs behind week-long autonomous coding runs, from flat agents with locking to optimistic concurrency to a planner/worker/judge hierarchy, letting hundreds of concurrent workers push to one branch on projects exceeding a million lines. | Cursor traces the coordination designs behind week-long autonomous coding runs, from flat agents with locking to optimistic concurrency to a planner/worker/judge hierarchy, letting hundreds of concurrent workers push to one branch on projects exceeding a million lines. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Cursor traces the coordination designs behind week-long autonomous coding runs, from flat agents with locking to optimistic concurrency to a planner/worker/judge hierarchy, letting hundreds of concurrent workers push to one bran... | Use Scaling Long-Running Autonomous Coding to choose an implementation surface for repeatable agent work. | Contextual source from cursor.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,413 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1413 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | practitioner-analysis | B | ok | https://cursor.com/blog/scaling-agents | Scaling long-running autonomous coding · Cursor | We've been experimenting with running coding agents autonomously for weeks at a time. | Wilson Lin | Cursor | html-meta | 2026-09-08T16:27:48 | ||||||||||||||
ale-0717 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | babysitter | https://github.com/a5c-ai/babysitter | external | github.com | Harness-agnostic orchestration framework that runs agent workflows as process-as-code with mandatory enforcement stops after every step, quality-convergence loops that re-run verify-and-refine until thresholds pass, human-approval breakpoints, and an immutable event-sourced journal for deterministic replay across 12 co... | Harness-agnostic orchestration framework that runs agent workflows as process-as-code with mandatory enforcement stops after every step, quality-convergence loops that re-run verify-and-refine until thresholds pass, human-approval breakpoints, and an immutable event-sourced journal for deterministic replay across 12 co... | Harness-agnostic orchestration framework that runs agent workflows as process-as-code with mandatory enforcement stops after every step, quality-convergence loops that re-run verify-and-refine until thresholds pass, human-approval breakpoints, and an immutable event-sourced journal for deterministic replay across 12 co... | Durable execution and replay are treated as first-class loop infrastructure. Harness-agnostic orchestration framework that runs agent workflows as process-as-code with mandatory enforcement stops after every step, quality-convergence loops that re-run verify-and-refine until thresholds pass, human-approval breakpoints,... | Use babysitter to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,414 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1414 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;verification;state;escalation;exit | builder | workflow | enabling | source-implementation | A | ok | https://github.com/a5c-ai/babysitter | GitHub - a5c-ai/babysitter: Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration · GitHub | Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration - a5c-ai/babysitter | a5c-ai/babysitter | GitHub | html-meta | a5c-ai/babysitter | 2026-09-08T16:27:48 | |||||||||||||
ale-0718 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | claude-code-merge-queue | https://github.com/funador/claude-code-merge-queue | external | github.com | Local FIFO merge queue that serializes parallel Claude Code agents landing on a shared codebase, with a machine-wide build lock and a landing gate that blocks the integration branch until a configured check command passes, wired into the WorktreeCreate hook. | Local FIFO merge queue that serializes parallel Claude Code agents landing on a shared codebase, with a machine-wide build lock and a landing gate that blocks the integration branch until a configured check command passes, wired into the WorktreeCreate hook. | Local FIFO merge queue that serializes parallel Claude Code agents landing on a shared codebase, with a machine-wide build lock and a landing gate that blocks the integration branch until a configured check command passes, wired into the WorktreeCreate hook. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Local FIFO merge queue that serializes parallel Claude Code agents landing on a shared codebase, with a machine-wide build lock and a landing gate that blocks the integration branch until a configured check command passes, wired... | Use claude-code-merge-queue to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,415 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1415 | Build | build | Choose runtimes, tools, and delegation surfaces. | intake | builder | workflow | enabling | source-implementation | A | ok | https://github.com/funador/claude-code-merge-queue | GitHub - funador/claude-code-merge-queue: Local merge queue for parallel Claude Code agents · GitHub | Local merge queue for parallel Claude Code agents. Contribute to funador/claude-code-merge-queue development by creating an account on GitHub. | funador/claude-code-merge-queue | GitHub | html-meta | funador/claude-code-merge-queue | 2026-09-08T16:27:48 | |||||||||||||
ale-0719 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Devin can now manage Devins | https://cognition.com/blog/devin-can-now-manage-devins | external | cognition.com | Cognition's multi-Devin architecture where a coordinator Devin scopes a task into pieces, delegates each to a managed Devin running in its own isolated VM with terminal, browser, and dev environment, monitors progress, resolves conflicts, compiles the results, and reads workers' full trajectories to learn what worked a... | Cognition's multi-Devin architecture where a coordinator Devin scopes a task into pieces, delegates each to a managed Devin running in its own isolated VM with terminal, browser, and dev environment, monitors progress, resolves conflicts, compiles the results, and reads workers' full trajectories to learn what worked a... | Cognition's multi-Devin architecture where a coordinator Devin scopes a task into pieces, delegates each to a managed Devin running in its own isolated VM with terminal, browser, and dev environment, monitors progress, resolves conflicts, compiles the results, and reads workers' full trajectories to learn what worked a... | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Cognition's multi-Devin architecture where a coordinator Devin scopes a task into pieces, delegates each to a managed Devin running in its own isolated VM with terminal, browser, and dev environment, monitors progress, resolves ... | Use Devin can now manage Devins to choose an implementation surface for repeatable agent work. | Contextual source from cognition.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,416 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1416 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | builder | workflow | enabling | practitioner-analysis | B | ok | https://cognition.com/blog/devin-can-now-manage-devins | Devin can now Manage Devins | Cognition | Devin can now break down large tasks and delegate them to a team of managed Devins, with each running in its own isolated VM in parallel. | The Cognition Team | 2026-03-19 | 2026 | cognition.com | html-meta | 2026-09-08T16:27:48 | ||||||||||||
ale-0720 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | pilotfish | https://github.com/Nanako0129/pilotfish | external | github.com | Multi-model orchestration layer for Claude Code where the frontier model plans, decides, and reviews while cheaper models execute volume work through six global subagent roles, with quality guarded by fresh-context verifier subagents rather than expensive models everywhere and graceful degradation when the frontier mod... | Multi-model orchestration layer for Claude Code where the frontier model plans, decides, and reviews while cheaper models execute volume work through six global subagent roles, with quality guarded by fresh-context verifier subagents rather than expensive models everywhere and graceful degradation when the frontier mod... | Multi-model orchestration layer for Claude Code where the frontier model plans, decides, and reviews while cheaper models execute volume work through six global subagent roles, with quality guarded by fresh-context verifier subagents rather than expensive models everywhere and graceful degradation when the frontier mod... | Verification is promoted from a final check to a loop-control signal. Multi-model orchestration layer for Claude Code where the frontier model plans, decides, and reviews while cheaper models execute volume work through six global subagent roles, with quality guarded by fresh-context verifier subagents rather than expe... | Use pilotfish to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,417 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1417 | Build | build | Choose runtimes, tools, and delegation surfaces. | context;delegation | builder | workflow | enabling | source-implementation | A | ok | https://github.com/Nanako0129/pilotfish | GitHub - Nanako0129/pilotfish: Multi-model orchestration layer for Claude Code — the frontier model plans, cheaper models execute, verification guards quality. One-prompt install. · GitHub | Multi-model orchestration layer for Claude Code — the frontier model plans, cheaper models execute, verification guards quality. One-prompt install. - Nanako0129/pilotfish | Nanako0129/pilotfish | GitHub | html-meta | Nanako0129/pilotfish | 2026-09-08T16:27:48 | |||||||||||||
ale-0721 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | fable-advisor | https://github.com/DannyMac180/fable-advisor | external | github.com | Claude Code plugin implementing an architect pattern where Fable 5 owns specs, decomposition, and verification while routing implementation to cheaper lanes (Grok 4.5 by default, GPT-5.6 via Codex CLI, or Sonnet/Opus fallback), with cross-vendor review and optional racing of two implementers on the same spec. | Claude Code plugin implementing an architect pattern where Fable 5 owns specs, decomposition, and verification while routing implementation to cheaper lanes (Grok 4.5 by default, GPT-5.6 via Codex CLI, or Sonnet/Opus fallback), with cross-vendor review and optional racing of two implementers on the same spec. | Claude Code plugin implementing an architect pattern where Fable 5 owns specs, decomposition, and verification while routing implementation to cheaper lanes (Grok 4.5 by default, GPT-5.6 via Codex CLI, or Sonnet/Opus fallback), with cross-vendor review and optional racing of two implementers on the same spec. | Verification is promoted from a final check to a loop-control signal. Claude Code plugin implementing an architect pattern where Fable 5 owns specs, decomposition, and verification while routing implementation to cheaper lanes (Grok 4.5 by default, GPT-5.6 via Codex CLI, or Sonnet/Opus fallback), with cross-vendor revi... | Use fable-advisor to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,418 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1418 | Build | build | Choose runtimes, tools, and delegation surfaces. | verification | builder | workflow | enabling | source-implementation | A | ok | https://github.com/DannyMac180/fable-advisor | GitHub - DannyMac180/fable-advisor: Claude Fable as an orchestrator for Opus, GPT and Grok · GitHub | Claude Fable as an orchestrator for Opus, GPT and Grok - DannyMac180/fable-advisor | DannyMac180/fable-advisor | GitHub | html-meta | DannyMac180/fable-advisor | 2026-09-08T16:27:48 | |||||||||||||
ale-0722 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | OpenTag | https://github.com/amplifthq/opentag | external | github.com | Turns an existing work thread into a governed agent work loop: @-mention a coding agent from Slack, GitHub, GitLab, Linear, Lark, Telegram, or Discord, and OpenTag curates the context and manages the run. | Turns an existing work thread into a governed agent work loop: @-mention a coding agent from Slack, GitHub, GitLab, Linear, Lark, Telegram, or Discord, and OpenTag curates the context and manages the run. | Turns an existing work thread into a governed agent work loop: @-mention a coding agent from Slack, GitHub, GitLab, Linear, Lark, Telegram, or Discord, and OpenTag curates the context and manages the run. | Context is managed as durable loop state rather than a single prompt payload. Turns an existing work thread into a governed agent work loop: @-mention a coding agent from Slack, GitHub, GitLab, Linear, Lark, Telegram, or Discord, and OpenTag curates the context and manages the run. | Use OpenTag to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,419 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1419 | Build | build | Choose runtimes, tools, and delegation surfaces. | context | builder | workflow | enabling | source-implementation | A | ok | https://github.com/amplifthq/opentag | GitHub - amplifthq/opentag: Mention any ACP coding agent from Slack, GitHub, GitLab, Linear, or Lark. OpenTag runs Claude Code, Codex, Cursor and more on your own machine, then replies in-thread with verified, evidence-backed results. · GitHub | Mention any ACP coding agent from Slack, GitHub, GitLab, Linear, or Lark. OpenTag runs Claude Code, Codex, Cursor and more on your own machine, then replies in-thread with verified, evidence-backed results. - amplifthq/opentag | amplifthq/opentag | GitHub | html-meta | amplifthq/opentag | 2026-09-08T16:27:48 | |||||||||||||
ale-0723 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | herdr | https://github.com/ogulcancelik/herdr | external | github.com | Single-binary terminal multiplexer purpose-built for running fleets of coding agents (Claude Code, Codex, Copilot CLI, Cursor Agent, and 15+ others) with per-agent state tracking. | Single-binary terminal multiplexer purpose-built for running fleets of coding agents (Claude Code, Codex, Copilot CLI, Cursor Agent, and 15+ others) with per-agent state tracking. | Single-binary terminal multiplexer purpose-built for running fleets of coding agents (Claude Code, Codex, Copilot CLI, Cursor Agent, and 15+ others) with per-agent state tracking. | State persistence is explicit enough for repeated runs and handoff. Single-binary terminal multiplexer purpose-built for running fleets of coding agents (Claude Code, Codex, Copilot CLI, Cursor Agent, and 15+ others) with per-agent state tracking. | Use herdr to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,420 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1420 | Build | build | Choose runtimes, tools, and delegation surfaces. | state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/herdrdev/herdr | GitHub - herdrdev/herdr: the runtime your coding agents live on · GitHub | the runtime your coding agents live on. Contribute to herdrdev/herdr development by creating an account on GitHub. | ogulcancelik/herdr | GitHub | html-meta | ogulcancelik/herdr | 2026-09-08T16:27:48 | |||||||||||||
ale-0724 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Orca | https://github.com/stablyai/orca | external | github.com | Open-source agent development environment for orchestrating a fleet of parallel coding agents (20+ backends) on your own subscriptions, with per-agent workspaces and review flow. | Open-source agent development environment for orchestrating a fleet of parallel coding agents (20+ backends) on your own subscriptions, with per-agent workspaces and review flow. | Open-source agent development environment for orchestrating a fleet of parallel coding agents (20+ backends) on your own subscriptions, with per-agent workspaces and review flow. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Open-source agent development environment for orchestrating a fleet of parallel coding agents (20+ backends) on your own subscriptions, with per-agent workspaces and review flow. | Use Orca to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,421 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1421 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace | builder | workflow | enabling | source-implementation | A | ok | https://github.com/stablyai/orca | GitHub - stablyai/orca: Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime. · GitHub | Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime. - stablyai/orca | stablyai/orca | GitHub | html-meta | stablyai/orca | 2026-09-08T16:27:48 | |||||||||||||
ale-0725 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Agentic Routing: The Harness-Native Data Flywheel | https://arxiv.org/abs/2607.11399 | external | arxiv.org | Argues model routing must live inside the execution harness rather than in single-turn cost-quality tradeoffs, making step-level model selections from execution state and logging each decision as structured telemetry that feeds back to improve both routers and models. | Argues model routing must live inside the execution harness rather than in single-turn cost-quality tradeoffs, making step-level model selections from execution state and logging each decision as structured telemetry that feeds back to improve both routers and models. | Argues model routing must live inside the execution harness rather than in single-turn cost-quality tradeoffs, making step-level model selections from execution state and logging each decision as structured telemetry that feeds back to improve both routers and models. | State persistence is explicit enough for repeated runs and handoff. Argues model routing must live inside the execution harness rather than in single-turn cost-quality tradeoffs, making step-level model selections from execution state and logging each decision as structured telemetry that feeds back to improve both rou... | Use Agentic Routing: The Harness-Native Data Flywheel to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.11399; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,422 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1422 | 2026-07-15 | Build | build | Choose runtimes, tools, and delegation surfaces. | state;budget | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.11399 | [2607.11399] Agentic Routing: The Harness-Native Data Flywheel | Large language model agents are increasingly executed not by a single model call, but by an execution harness that manages observation, context, control, action, state, and verification. At the same time, frontier and open models are becoming structurally specialized: a model that is strong at code editing, long-contex... | Xinchen Liu; Hang Zhou; Yingjie Zong; Yuchuan Tian; Liuyang Song; Shuo Zhang; Yulong Li; Wei He; Mengyu Zheng; Runke Liu; Siyang Cheng; Xiang Kuang; Hailin Hu; Kai Han; Yunhe Wang | 2026-07-13 | 2026 | arXiv | arXiv | Code: https://github.com/opensquilla/opensquilla | cs.CL | arxiv-api | 2607.11399 | 2026-09-08T16:27:48 | |||||||
ale-0726 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | A Formal Hierarchical Architecture for Agentic Orchestration with Stack-Based Execution | https://arxiv.org/abs/2607.11138 | external | arxiv.org | Formal orchestration architecture that runs agent workflows on a call stack with lazy capability discovery, giving multi-agent loops explicit, inspectable control flow instead of implicit prompt-driven handoffs. | Formal orchestration architecture that runs agent workflows on a call stack with lazy capability discovery, giving multi-agent loops explicit, inspectable control flow instead of implicit prompt-driven handoffs. | Formal orchestration architecture that runs agent workflows on a call stack with lazy capability discovery, giving multi-agent loops explicit, inspectable control flow instead of implicit prompt-driven handoffs. | The work separates roles across agents, verifiers, or orchestration layers. Formal orchestration architecture that runs agent workflows on a call stack with lazy capability discovery, giving multi-agent loops explicit, inspectable control flow instead of implicit prompt-driven handoffs. | Use A Formal Hierarchical Architecture for Agentic Orchestration with Stack-Based Execution to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.11138; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,423 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1423 | 2026-07-15 | Build | build | Choose runtimes, tools, and delegation surfaces. | intake;delegation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.11138 | [2607.11138] A Formal Hierarchical Architecture for Agentic Orchestration with Stack-Based Execution and Lazy Discovery | The rapid expansion of capabilities in Large Language Model (LLM) agents has exposed a critical architectural bottleneck: when agents are given access to a flat, monolithic registry of tools, the model must evaluate hundreds or thousands of options simultaneously. This leads to decision-space explosion, context window ... | Prashant Devadiga; Abhishek; Adithya Mishra; Alok Singh; Amisha Sinha; Asit Desai; Gaurang Dahad; Harshit Bhushan; Mandati Pramod Reddy; Prakhar Gupta; Rupesh Patil; Siddhi Behere | 2026-07-13 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2607.11138 | 2026-09-08T16:27:48 | ||||||||
ale-0727 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | The Honest Quorum Problem: Epistemic Byzantine Fault Tolerance for Agentic Infrastructure | https://arxiv.org/abs/2607.16109 | external | arxiv.org | Formalizes how protocol-compliant reasoning agents can still agree on a semantically invalid transition, especially when they share models, prompts, or tools; derives separate safety and liveness budgets and shows that adding agents helps only when measured error correlation falls. | Formalizes how protocol-compliant reasoning agents can still agree on a semantically invalid transition, especially when they share models, prompts, or tools; derives separate safety and liveness budgets and shows that adding agents helps only when measured error correlation falls. | Formalizes how protocol-compliant reasoning agents can still agree on a semantically invalid transition, especially when they share models, prompts, or tools; derives separate safety and liveness budgets and shows that adding agents helps only when measured error correlation falls. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Formalizes how protocol-compliant reasoning agents can still agree on a semantically invalid transition, especially when they share models, prompts, or tools; derives separate safety and liveness budgets and shows that adding ag... | Use The Honest Quorum Problem: Epistemic Byzantine Fault Tolerance for Agentic Infrastructure to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.16109; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,424 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1424 | 2026-07-20 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;budget | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.16109 | [2607.16109] The Honest Quorum Problem: Epistemic Byzantine Fault Tolerance for Agentic Infrastructure | State machine replication (SMR) and Byzantine fault-tolerant (BFT) consensus guarantee agreement despite a bounded number of arbitrary, colluding faulty participants. However, these guarantees rely on participants outside this set correctly executing the protocol's transition semantics. Agentic validators expose a weak... | Jun He; Deying Yu | 2026-07-17 | 2026 | arXiv | arXiv | 26 pages, 2 figures, 5 tables | cs.LG | arxiv-api | 2607.16109 | 2026-09-08T16:27:48 | |||||||
ale-0728 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Docs | 📚 | Graph-based agent workflows | https://adk.dev/graphs/ | external | adk.dev | Official ADK 2.0 graph runtime for explicit, deterministic workflows that mix agents, tools, and code with branching, state, human input, and bounded cycles. | Official ADK 2.0 graph runtime for explicit, deterministic workflows that mix agents, tools, and code with branching, state, human input, and bounded cycles. | Official ADK 2.0 graph runtime for explicit, deterministic workflows that mix agents, tools, and code with branching, state, human input, and bounded cycles. | Primary-source operational guidance rather than commentary. Official ADK 2.0 graph runtime for explicit, deterministic workflows that mix agents, tools, and code with branching, state, human input, and bounded cycles. | Use Graph-based agent workflows to choose an implementation surface for repeatable agent work. | Primary official documentation from adk.dev; use it for current product or standard behavior. | high | README.md | 1,425 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1425 | 2026-07-17 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;state;escalation | builder | workflow | enabling | official-documentation | A | ok | https://adk.dev/graphs/ | Graph-based agent workflows - Agent Development Kit (ADK) Agent Development Kit (ADK) | Build powerful multi-agent systems with Agent Development Kit (ADK) | Google Agent Development Kit | Google Agent Development Kit | Google | primary-page | 2026-09-08T16:27:48 | ||||||||||||
ale-0729 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Docs | 📚 | Flows | https://docs.crewai.com/en/concepts/flows | external | docs.crewai.com | Official event-driven workflow layer with typed state, branching and loops, persistence decorators, and resume or fork operations for long-running executions. | Official event-driven workflow layer with typed state, branching and loops, persistence decorators, and resume or fork operations for long-running executions. | Official event-driven workflow layer with typed state, branching and loops, persistence decorators, and resume or fork operations for long-running executions. | Primary-source operational guidance rather than commentary. Official event-driven workflow layer with typed state, branching and loops, persistence decorators, and resume or fork operations for long-running executions. | Use Flows to choose an implementation surface for repeatable agent work. | Primary official documentation from docs.crewai.com; use it for current product or standard behavior. | high | README.md | 1,426 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1426 | 2026-07-17 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;state | builder | workflow | enabling | official-documentation | A | ok | https://docs.crewai.com/v1.15.20/en/concepts/flows | Flows - CrewAI | Learn how to create and manage AI workflows using CrewAI Flows. | CrewAI | CrewAI | CrewAI | primary-page | 2026-09-08T16:27:48 | ||||||||||||
ale-0730 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Docs | 📚 | Graph | https://strandsagents.com/docs/user-guide/concepts/multi-agent/graph/ | external | strandsagents.com | Official deterministic graph pattern for multi-agent dependencies and cycles, with shared execution state and explicit execution limits. | Official deterministic graph pattern for multi-agent dependencies and cycles, with shared execution state and explicit execution limits. | Official deterministic graph pattern for multi-agent dependencies and cycles, with shared execution state and explicit execution limits. | Primary-source operational guidance rather than commentary. Official deterministic graph pattern for multi-agent dependencies and cycles, with shared execution state and explicit execution limits. | Use Graph to choose an implementation surface for repeatable agent work. | Primary official documentation from strandsagents.com; use it for current product or standard behavior. | high | README.md | 1,427 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1427 | 2026-07-17 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | official-documentation | A | ok | https://strandsagents.com/docs/user-guide/concepts/multi-agent/graph/ | Graph Multi-Agent Pattern | Strands Agents | Orchestrate multi-agent systems with the Strands Graph pattern: deterministic DAG or cyclic execution, conditional edges, and nested agents. | Strands Agents | Strands Agents | Strands Agents | primary-page | 2026-09-08T16:27:48 | ||||||||||||
ale-0731 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Towards a Science of Scaling Agent Systems | https://arxiv.org/abs/2512.08296 | external | arxiv.org | Controls 260 configurations across six agentic benchmarks to show how coordination, model capability, task parallelism, and tool load shape performance and error amplification. | Controls 260 configurations across six agentic benchmarks to show how coordination, model capability, task parallelism, and tool load shape performance and error amplification. | Controls 260 configurations across six agentic benchmarks to show how coordination, model capability, task parallelism, and tool load shape performance and error amplification. | The work turns loop quality into a measurable task or score. Controls 260 configurations across six agentic benchmarks to show how coordination, model capability, task parallelism, and tool load shape performance and error amplification. | Use Towards a Science of Scaling Agent Systems to choose an implementation surface for repeatable agent work. | Research source arXiv:2512.08296; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,431 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1431 | 2026-07-18 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;verification | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2512.08296 | [2512.08296] Towards a Science of Scaling Agent Systems | Agents, language model-based systems capable of reasoning, planning, and acting are widely adopted in real-world tasks, yet how their performance changes as these systems scale across key dimensions remains underexplored. We introduce quantitative scaling principles for agent systems as a predictive model, capturing ho... | Yubin Kim; Ken Gu; Chanwoo Park; Chunjong Park; Samuel Schmidgall; A. Ali Heydari; Yao Yan; Zhihan Zhang; Yuchen Zhuang; Yun Liu; Mark Malhotra; Paul Pu Liang; Hae Won Park; Yuzhe Yang; Xuhai Xu; Yilun Du; Shwetak Patel; Tim Althoff; Daniel McDuff; Xin Liu | 2025-12-09 | 2025 | arXiv | arXiv | cs.AI | arxiv-api | 2512.08296 | 2026-09-08T16:27:48 | ||||||||
ale-0732 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Amp: From Agent to Agent | https://ampcode.com/news/from-agent-to-agent | external | ampcode.com | Amp on agent-to-agent handoffs: one agent delegating bounded work to another with results returned to the parent thread, formalizing delegation as a first-class platform primitive. | Amp on agent-to-agent handoffs: one agent delegating bounded work to another with results returned to the parent thread, formalizing delegation as a first-class platform primitive. | Amp on agent-to-agent handoffs: one agent delegating bounded work to another with results returned to the parent thread, formalizing delegation as a first-class platform primitive. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Amp on agent-to-agent handoffs: one agent delegating bounded work to another with results returned to the parent thread, formalizing delegation as a first-class platform primitive. | Use Amp: From Agent to Agent to choose an implementation surface for repeatable agent work. | Contextual source from ampcode.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,432 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1432 | 2026-07-22 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | builder | workflow | enabling | practitioner-analysis | B | ok | https://ampcode.com/news/from-agent-to-agent | From Agent to Agent - Amp | Agents can spawn other agents, message them, and exchange files across Amp threads. | Amp | 2026-07-17 | 2026 | ampcode.com | html-meta | 2026-09-08T16:27:48 | |||||||||||
ale-0733 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Cursor: Agent Swarms and the New Model Economics | https://cursor.com/blog/agent-swarm-model-economics | external | cursor.com | Cursor on the economics of agent swarms: how parallel fleets change the cost calculus of model selection, and what coordination overhead does to effective throughput per dollar. | Cursor on the economics of agent swarms: how parallel fleets change the cost calculus of model selection, and what coordination overhead does to effective throughput per dollar. | Cursor on the economics of agent swarms: how parallel fleets change the cost calculus of model selection, and what coordination overhead does to effective throughput per dollar. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Cursor on the economics of agent swarms: how parallel fleets change the cost calculus of model selection, and what coordination overhead does to effective throughput per dollar. | Use Cursor: Agent Swarms and the New Model Economics to choose an implementation surface for repeatable agent work. | Contextual source from cursor.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,433 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1433 | 2026-07-22 | Build | build | Choose runtimes, tools, and delegation surfaces. | budget | builder | workflow | enabling | practitioner-analysis | B | ok | https://cursor.com/blog/agent-swarm-model-economics | Agent swarms and the new model economics · Cursor | We compared old and new agent swarms building SQLite from scratch and found that better coordination delivers similar quality at a fraction of the cost. | Wilson Lin | Cursor | html-meta | 2026-09-08T16:27:48 | |||||||||||||
ale-0734 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Amp: Meet Puck | https://ampcode.com/news/meet-puck | external | ampcode.com | Amp's conversational coordinator that spawns agents, investigates issues, organizes threads, and coordinates multiple agent tasks across projects through natural-language commands. | Amp's conversational coordinator that spawns agents, investigates issues, organizes threads, and coordinates multiple agent tasks across projects through natural-language commands. | Amp's conversational coordinator that spawns agents, investigates issues, organizes threads, and coordinates multiple agent tasks across projects through natural-language commands. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Amp's conversational coordinator that spawns agents, investigates issues, organizes threads, and coordinates multiple agent tasks across projects through natural-language commands. | Use Amp: Meet Puck to choose an implementation surface for repeatable agent work. | Contextual source from ampcode.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,434 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1434 | 2026-07-22 | Build | build | Choose runtimes, tools, and delegation surfaces. | intake | builder | workflow | enabling | practitioner-analysis | B | ok | https://ampcode.com/news/meet-puck | Meet Puck - Amp | Puck is your new meta-agent, your Amp helper | Amp | 2026-07-20 | 2026 | ampcode.com | html-meta | 2026-09-08T16:27:48 | |||||||||||
ale-0735 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Warren | https://github.com/jayminwest/warren | external | github.com | Control plane for coding agents that operate in isolation, styled as a Coolify for agents: deploy, monitor, and manage long-running agent workloads on your own infrastructure. | Control plane for coding agents that operate in isolation, styled as a Coolify for agents: deploy, monitor, and manage long-running agent workloads on your own infrastructure. | Control plane for coding agents that operate in isolation, styled as a Coolify for agents: deploy, monitor, and manage long-running agent workloads on your own infrastructure. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Control plane for coding agents that operate in isolation, styled as a Coolify for agents: deploy, monitor, and manage long-running agent workloads on your own infrastructure. | Use Warren to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,435 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1435 | 2026-07-22 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/jayminwest/warren | GitHub - jayminwest/warren: Run coding agents like infrastructure, not terminal sessions. Warren manages isolation, lifecycle, spend, recovery, and Git delivery on compute you control. · GitHub | Run coding agents like infrastructure, not terminal sessions. Warren manages isolation, lifecycle, spend, recovery, and Git delivery on compute you control. - jayminwest/warren | jayminwest/warren | GitHub | html-meta | jayminwest/warren | 2026-09-08T16:27:48 | ||||||||||||
ale-0736 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | agent-talk | https://github.com/xhluca/agent-talk | external | github.com | Enables coding agents to work together by giving them a shared communication channel, so parallel agents can coordinate instead of colliding. | Enables coding agents to work together by giving them a shared communication channel, so parallel agents can coordinate instead of colliding. | Enables coding agents to work together by giving them a shared communication channel, so parallel agents can coordinate instead of colliding. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Enables coding agents to work together by giving them a shared communication channel, so parallel agents can coordinate instead of colliding. | Use agent-talk to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,436 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1436 | 2026-07-22 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/xhluca/agent-talk | GitHub - xhluca/agent-talk: agent-talk enables any coding agent to talk to each other · GitHub | agent-talk enables any coding agent to talk to each other - xhluca/agent-talk | xhluca/agent-talk | GitHub | html-meta | xhluca/agent-talk | 2026-09-08T16:27:48 | ||||||||||||
ale-0737 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | codex-model-routing-team | https://github.com/zjp1997720/codex-model-routing-team | external | github.com | Runs background Codex workers under a lead agent that verifies their output before it lands, a small-team pattern for supervised parallel delegation. | Runs background Codex workers under a lead agent that verifies their output before it lands, a small-team pattern for supervised parallel delegation. | Runs background Codex workers under a lead agent that verifies their output before it lands, a small-team pattern for supervised parallel delegation. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Runs background Codex workers under a lead agent that verifies their output before it lands, a small-team pattern for supervised parallel delegation. | Use codex-model-routing-team to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,437 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1437 | 2026-07-22 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;verification | builder | workflow | enabling | source-implementation | A | ok | https://github.com/zjp1997720/codex-model-routing-team | GitHub - zjp1997720/codex-model-routing-team: Route complex Codex tasks to model-specific background workers with bounded concurrency and lead-agent verification. · GitHub | Route complex Codex tasks to model-specific background workers with bounded concurrency and lead-agent verification. - zjp1997720/codex-model-routing-team | zjp1997720/codex-model-routing-team | GitHub | html-meta | zjp1997720/codex-model-routing-team | 2026-09-08T16:27:48 | ||||||||||||
ale-0738 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Agent Orchestrator | https://github.com/AgentWrapper/agent-orchestrator | external | github.com | Agent IDE and orchestrator for managing fleets of coding agents, with worktree isolation and coordination so many agents can work a backlog in parallel from one control surface. | Agent IDE and orchestrator for managing fleets of coding agents, with worktree isolation and coordination so many agents can work a backlog in parallel from one control surface. | Agent IDE and orchestrator for managing fleets of coding agents, with worktree isolation and coordination so many agents can work a backlog in parallel from one control surface. | Workspace isolation is part of the loop design, not an afterthought. Agent IDE and orchestrator for managing fleets of coding agents, with worktree isolation and coordination so many agents can work a backlog in parallel from one control surface. | Use Agent Orchestrator to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,438 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1438 | 2026-07-23 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;delegation | builder | workflow | enabling | source-implementation | A | ok | https://github.com/Untrivial-ai/agent-orchestrator | GitHub - Untrivial-ai/agent-orchestrator: Agent IDE that enables you to manage fleets of coding agents. It comes with an agentic orchestrator that plans tasks, spawns agents, and autonomously handles CI fixes, merge conflicts, and code reviews. · GitHub | Agent IDE that enables you to manage fleets of coding agents. It comes with an agentic orchestrator that plans tasks, spawns agents, and autonomously handles CI fixes, merge conflicts, and code reviews. - Untrivial-ai/agent-orchestrator | AgentWrapper/agent-orchestrator | GitHub | html-meta | AgentWrapper/agent-orchestrator | 2026-09-08T16:27:48 | ||||||||||||
ale-0739 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Buzz | https://github.com/block/buzz | external | github.com | Self-hostable workspace from Block where humans and AI agents share the same rooms, giving a fleet of long-running agents a common coordination surface instead of isolated one-off sessions. | Self-hostable workspace from Block where humans and AI agents share the same rooms, giving a fleet of long-running agents a common coordination surface instead of isolated one-off sessions. | Self-hostable workspace from Block where humans and AI agents share the same rooms, giving a fleet of long-running agents a common coordination surface instead of isolated one-off sessions. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Self-hostable workspace from Block where humans and AI agents share the same rooms, giving a fleet of long-running agents a common coordination surface instead of isolated one-off sessions. | Use Buzz to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,439 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1439 | 2026-07-24 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace | builder | workflow | enabling | source-implementation | A | ok | https://github.com/block/buzz | GitHub - block/buzz: A hive mind communication platform · GitHub | A hive mind communication platform. Contribute to block/buzz development by creating an account on GitHub. | block/buzz | GitHub | html-meta | block/buzz | 2026-09-08T16:27:48 | ||||||||||||
ale-0740 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | open-kritt | https://github.com/Kritt-ai/open-kritt | external | github.com | AGPL-3.0 security-research platform that decomposes vulnerability hunting into small well-defined tasks, fans them out across parallel tool-enabled agents in disposable Docker sandboxes, then closes the loop with post-processing verification scripts, proof-of-concept generation, de-duplication, and severity ranking via... | AGPL-3.0 security-research platform that decomposes vulnerability hunting into small well-defined tasks, fans them out across parallel tool-enabled agents in disposable Docker sandboxes, then closes the loop with post-processing verification scripts, proof-of-concept generation, de-duplication, and severity ranking via... | AGPL-3.0 security-research platform that decomposes vulnerability hunting into small well-defined tasks, fans them out across parallel tool-enabled agents in disposable Docker sandboxes, then closes the loop with post-processing verification scripts, proof-of-concept generation, de-duplication, and severity ranking via... | Verification is promoted from a final check to a loop-control signal. AGPL-3.0 security-research platform that decomposes vulnerability hunting into small well-defined tasks, fans them out across parallel tool-enabled agents in disposable Docker sandboxes, then closes the loop with post-processing verification scripts,... | Use open-kritt to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,440 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1440 | 2026-07-24 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;verification | builder | workflow | enabling | source-implementation | A | ok | https://github.com/Kritt-ai/open-kritt | GitHub - Kritt-ai/open-kritt: Open-source, self-hosted AI vulnerability research tool that orchestrates agents to find and validate security issues in code. · GitHub | Open-source, self-hosted AI vulnerability research tool that orchestrates agents to find and validate security issues in code. - Kritt-ai/open-kritt | Kritt-ai/open-kritt | GitHub | html-meta | Kritt-ai/open-kritt | 2026-09-08T16:27:48 | ||||||||||||
ale-0741 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | BossConsole | https://github.com/risa-labs-inc/BossConsole | external | github.com | Native JVM operator's console (Kotlin Multiplatform, explicitly not Electron) that runs Claude Code, Codex, Gemini CLI, and OpenCode side-by-side with a shared browser, terminal, editor, secrets manager, and ~100 MCP tools, adding per-tool RBAC governance, kill-switches, E2E-encrypted session sharing, and daemonized pe... | Native JVM operator's console (Kotlin Multiplatform, explicitly not Electron) that runs Claude Code, Codex, Gemini CLI, and OpenCode side-by-side with a shared browser, terminal, editor, secrets manager, and ~100 MCP tools, adding per-tool RBAC governance, kill-switches, E2E-encrypted session sharing, and daemonized pe... | Native JVM operator's console (Kotlin Multiplatform, explicitly not Electron) that runs Claude Code, Codex, Gemini CLI, and OpenCode side-by-side with a shared browser, terminal, editor, secrets manager, and ~100 MCP tools, adding per-tool RBAC governance, kill-switches, E2E-encrypted session sharing, and daemonized pe... | The work separates roles across agents, verifiers, or orchestration layers. Native JVM operator's console (Kotlin Multiplatform, explicitly not Electron) that runs Claude Code, Codex, Gemini CLI, and OpenCode side-by-side with a shared browser, terminal, editor, secrets manager, and ~100 MCP tools, adding per-tool RBAC... | Use BossConsole to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,441 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1441 | 2026-07-24 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;delegation;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/risa-labs-inc/BossConsole | GitHub - risa-labs-inc/BossConsole: Open-source, multi-platform harness for AI agents - a native, multi-threaded operator's console (JVM, not Electron) to run Claude Code, Codex, Gemini or OpenCode with a real browser, terminal, editor, secrets & 100+ MCP tools. Built for enterprises, science & research. · GitHub | Open-source, multi-platform harness for AI agents - a native, multi-threaded operator's console (JVM, not Electron) to run Claude Code, Codex, Gemini or OpenCode with a real browser, terminal, editor, secrets & 100+ MCP tools. Built for enterprises, science & research. - risa-labs-inc/BossConsole | risa-labs-inc/BossConsole | GitHub | html-meta | risa-labs-inc/BossConsole | 2026-09-08T16:27:48 | ||||||||||||
ale-0742 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Fractal | https://github.com/plasma-ai/fractal | external | github.com | Arranges autonomous agent loops into a tree: each node iterates toward a goal in its own Git worktree and spawns child loop-nodes for separable subtasks, bounded by hard caps on iterations, depth, children, cost, and time, with run metadata in a local SQLite database and a live TUI for operator steering. | Arranges autonomous agent loops into a tree: each node iterates toward a goal in its own Git worktree and spawns child loop-nodes for separable subtasks, bounded by hard caps on iterations, depth, children, cost, and time, with run metadata in a local SQLite database and a live TUI for operator steering. | Arranges autonomous agent loops into a tree: each node iterates toward a goal in its own Git worktree and spawns child loop-nodes for separable subtasks, bounded by hard caps on iterations, depth, children, cost, and time, with run metadata in a local SQLite database and a live TUI for operator steering. | Workspace isolation is part of the loop design, not an afterthought. Arranges autonomous agent loops into a tree: each node iterates toward a goal in its own Git worktree and spawns child loop-nodes for separable subtasks, bounded by hard caps on iterations, depth, children, cost, and time, with run metadata in a local... | Use Fractal to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,442 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1442 | 2026-07-25 | Build | build | Choose runtimes, tools, and delegation surfaces. | objective;workspace;budget | builder | workflow | enabling | source-implementation | A | ok | https://github.com/plasma-ai/fractal | GitHub - plasma-ai/fractal: Hierarchical agent loops with recursive self-organization. · GitHub | Hierarchical agent loops with recursive self-organization. - plasma-ai/fractal | plasma-ai/fractal | GitHub | html-meta | plasma-ai/fractal | 2026-09-08T16:27:48 | ||||||||||||
ale-0743 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | SpecBox: Speculative Sandbox Scheduling for Efficient LLM Agent Serving | https://arxiv.org/abs/2607.23933 | external | arxiv.org | Serving infrastructure for MCP-based agent loops, where disaggregated sandboxes force a choice between persistent reservations that waste memory at scale and lazy instantiation that inflicts cold-start penalties on multi-tenant multi-turn workloads. | Serving infrastructure for MCP-based agent loops, where disaggregated sandboxes force a choice between persistent reservations that waste memory at scale and lazy instantiation that inflicts cold-start penalties on multi-tenant multi-turn workloads. | Serving infrastructure for MCP-based agent loops, where disaggregated sandboxes force a choice between persistent reservations that waste memory at scale and lazy instantiation that inflicts cold-start penalties on multi-tenant multi-turn workloads. | The trigger or cadence is explicit, making the workflow recurring rather than one-off. Serving infrastructure for MCP-based agent loops, where disaggregated sandboxes force a choice between persistent reservations that waste memory at scale and lazy instantiation that inflicts cold-start penalties on multi-tenant multi... | Use SpecBox: Speculative Sandbox Scheduling for Efficient LLM Agent Serving to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.23933; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,443 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1443 | 2026-07-28 | Build | build | Choose runtimes, tools, and delegation surfaces. | trigger;workspace;context;state | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.23933 | [2607.23933] SpecBox: Speculative Sandbox Scheduling for Efficient LLM Agent Serving | As LLM agents increasingly rely on the Model Context Protocol (MCP) to invoke isolated external sandboxes, disaggregated sandbox deployment introduces a fundamental tension between resource utilization and interactive tail latency. Persistent long-lived sandbox reservations incur excessive memory overhead at scale, whi... | Yihui Zhang; Tianyu Wo; Jinghao Wang; Xiaoyang Sun; Menghao Zhang; Cangzhou Yuan; Li Li; Chunming Hu; Albert Y. Zomaya; Renyu Yang | 2026-07-27 | 2026 | arXiv | arXiv | cs.DC | arxiv-api | 2607.23933 | 2026-09-08T16:27:48 | ||||||||
ale-0744 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | A Comparative Study of MCP and A2A for Inter-Agent Coordination in LLM-Based Systems | https://arxiv.org/abs/2607.23884 | external | arxiv.org | Implementation-grounded comparison rather than a survey: the same software-engineering task built twice, once on MCP and once on A2A, evaluated against requirements drawn from prior literature and industry partner discussions -- agent discoverability, multi-part messaging, multi-turn conversations, asynchronous communi... | Implementation-grounded comparison rather than a survey: the same software-engineering task built twice, once on MCP and once on A2A, evaluated against requirements drawn from prior literature and industry partner discussions -- agent discoverability, multi-part messaging, multi-turn conversations, asynchronous communi... | Implementation-grounded comparison rather than a survey: the same software-engineering task built twice, once on MCP and once on A2A, evaluated against requirements drawn from prior literature and industry partner discussions -- agent discoverability, multi-part messaging, multi-turn conversations, asynchronous communi... | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Implementation-grounded comparison rather than a survey: the same software-engineering task built twice, once on MCP and once on A2A, evaluated against requirements drawn from prior literature and industry partner discussions --... | Use A Comparative Study of MCP and A2A for Inter-Agent Coordination in LLM-Based Systems to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.23884; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,444 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1444 | 2026-07-28 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.23884 | [2607.23884] A Comparative Study of MCP and A2A for Inter-Agent Coordination in LLM-Based Systems | Recent industry practice has seen the rapid emergence of agentic systems composed of heterogeneous, tool- and LLM-mediated agent components, raising practical questions about inter-agent coordination and protocol design. This paper presents an implementation-grounded comparison of the Model Context Protocol (MCP) and t... | Ionut Predoaia; Tuong Manh Vu; Konstantinos Barmpis; Dimitris Kolovos; Antonio García-Domínguez | 2026-07-26 | 2026 | arXiv | arXiv | 18 pages | cs.MA | arxiv-api | 2607.23884 | 2026-09-08T16:27:48 | |||||||
ale-0745 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | AgentENV | https://github.com/kvcache-ai/AgentENV | external | github.com | Open-sourced 2026-07-27 by the Kimi/kvcache-ai team as the environment substrate behind Kimi K3's agentic RL training. | Open-sourced 2026-07-27 by the Kimi/kvcache-ai team as the environment substrate behind Kimi K3's agentic RL training. | Open-sourced 2026-07-27 by the Kimi/kvcache-ai team as the environment substrate behind Kimi K3's agentic RL training. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Open-sourced 2026-07-27 by the Kimi/kvcache-ai team as the environment substrate behind Kimi K3's agentic RL training. | Use AgentENV to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,445 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1445 | 2026-07-28 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/kvcache-ai/AgentENV | GitHub - kvcache-ai/AgentENV: AgentENV (AENV) is a distributed platform for running agent environments at scale. · GitHub | AgentENV (AENV) is a distributed platform for running agent environments at scale. - kvcache-ai/AgentENV | kvcache-ai/AgentENV | GitHub | html-meta | kvcache-ai/AgentENV | 2026-09-08T16:27:48 | ||||||||||||
ale-0746 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Ruflo | https://github.com/ruvnet/ruflo | external | github.com | Agent meta-harness for Claude Code and Codex, built on the premise that an agent is a model plus a harness: it supplies tools, memory, loops, sandboxes, and controls so agents self-organize into swarms, learn across tasks, and persist state between runs. Formerly Claude Flow. | Agent meta-harness for Claude Code and Codex, built on the premise that an agent is a model plus a harness: it supplies tools, memory, loops, sandboxes, and controls so agents self-organize into swarms, learn across tasks, and persist state between runs. Formerly Claude Flow. | Agent meta-harness for Claude Code and Codex, built on the premise that an agent is a model plus a harness: it supplies tools, memory, loops, sandboxes, and controls so agents self-organize into swarms, learn across tasks, and persist state between runs. Formerly Claude Flow. | Persistent memory is treated as an external runtime artifact. Agent meta-harness for Claude Code and Codex, built on the premise that an agent is a model plus a harness: it supplies tools, memory, loops, sandboxes, and controls so agents self-organize into swarms, learn across tasks, and persist state between runs. For... | Use Ruflo to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,446 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1446 | 2026-07-28 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;context;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/ruvnet/ruflo | GitHub - ruvnet/ruflo: 🌊 The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated · GitHub | 🌊 The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated - ruvnet/ruflo | ruvnet/ruflo | GitHub | html-meta | ruvnet/ruflo | 2026-09-08T16:27:48 | ||||||||||||
ale-0747 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Toward an Organizational Science of Multi-Agent LLM Systems: Decoupling Who, How, and Which Algorithm | https://arxiv.org/abs/2607.25446 | external | arxiv.org | IMACS separates multi-agent systems into organizational structure, coordination mechanism, and collaboration protocol, then shows a contextual-bandit meta-protocol picking per task beats any fixed protocol on quality-cost tradeoffs. The decoupling makes previously incomparable multi-agent results comparable. | IMACS separates multi-agent systems into organizational structure, coordination mechanism, and collaboration protocol, then shows a contextual-bandit meta-protocol picking per task beats any fixed protocol on quality-cost tradeoffs. The decoupling makes previously incomparable multi-agent results comparable. | IMACS separates multi-agent systems into organizational structure, coordination mechanism, and collaboration protocol, then shows a contextual-bandit meta-protocol picking per task beats any fixed protocol on quality-cost tradeoffs. The decoupling makes previously incomparable multi-agent results comparable. | The work separates roles across agents, verifiers, or orchestration layers. IMACS separates multi-agent systems into organizational structure, coordination mechanism, and collaboration protocol, then shows a contextual-bandit meta-protocol picking per task beats any fixed protocol on quality-cost tradeoffs. The decoupl... | Use Toward an Organizational Science of Multi-Agent LLM Systems: Decoupling Who, How, and Which Algorithm to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.25446; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,447 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1447 | 2026-07-30 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;budget | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.25446 | [2607.25446] Toward an Organizational Science of Multi-Agent LLM Systems: Decoupling Who, How, and Which Algorithm | Multi-agent frameworks built on large language models (LLMs) routinely entangle three logically distinct concerns: who is on the team (organization), how members align (coordination), and which algorithm fuses their work (collaboration protocol). IMACS (Intelligent Multi-Agent Collaboration System) separates the three ... | Huan Chen; Xiang Song; Jian Jin; Pan Ren; Liang-Jie Zhang | 2026-07-28 | 2026 | arXiv | arXiv | 8 pages, 2 figures | cs.AI | arxiv-api | 2607.25446 | 2026-09-08T16:27:48 | |||||||
ale-0748 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Matryoshka Agent: Unfolding Sub-Agents for Long-Horizon Machine Learning Engineering | https://arxiv.org/abs/2607.25090 | external | arxiv.org | Hierarchical orchestrator/sub-agent decomposition for long-horizon ML engineering that lets a 4B model reach performance comparable to much larger ones, with relative gains up to 36.7%. Good evidence that delegation structure substitutes for model scale on long tasks. | Hierarchical orchestrator/sub-agent decomposition for long-horizon ML engineering that lets a 4B model reach performance comparable to much larger ones, with relative gains up to 36.7%. Good evidence that delegation structure substitutes for model scale on long tasks. | Hierarchical orchestrator/sub-agent decomposition for long-horizon ML engineering that lets a 4B model reach performance comparable to much larger ones, with relative gains up to 36.7%. Good evidence that delegation structure substitutes for model scale on long tasks. | Orchestration and control flow are made explicit and inspectable. Hierarchical orchestrator/sub-agent decomposition for long-horizon ML engineering that lets a 4B model reach performance comparable to much larger ones, with relative gains up to 36.7%. Good evidence that delegation structure substitutes for model scale ... | Use Matryoshka Agent: Unfolding Sub-Agents for Long-Horizon Machine Learning Engineering to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.25090; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,448 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1448 | 2026-07-30 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.25090 | [2607.25090] Matryoshka Agent: Unfolding Sub-Agents for Long-Horizon Machine Learning Engineering | Machine learning engineering (MLE) tasks require long-horizon decision making over iterative solution debugging and refinement, under expensive and feedback-driven environment interactions. Developing and training a monolithic agent for such tasks is fundamentally challenging, as it must simultaneously manage extremely... | Rushi Qiang; Changhao Li; Haotian Sun; Yuchen Zhuang; Chao Zhang; Bo Dai | 2026-07-27 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2607.25090 | 2026-09-08T16:27:48 | ||||||||
ale-0749 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Multi-Agent Debate Strategies: Survey, Taxonomy, and Challenges | https://arxiv.org/abs/2607.26212 | external | arxiv.org | Systematic review of 141 multi-agent debate studies with a three-dimensional taxonomy over participants, interaction mechanisms, and agreement protocols, concluding the field has converged prematurely on one conventional design. Useful map for anyone choosing a consensus mechanism for a delegation loop. | Systematic review of 141 multi-agent debate studies with a three-dimensional taxonomy over participants, interaction mechanisms, and agreement protocols, concluding the field has converged prematurely on one conventional design. Useful map for anyone choosing a consensus mechanism for a delegation loop. | Systematic review of 141 multi-agent debate studies with a three-dimensional taxonomy over participants, interaction mechanisms, and agreement protocols, concluding the field has converged prematurely on one conventional design. Useful map for anyone choosing a consensus mechanism for a delegation loop. | The work separates roles across agents, verifiers, or orchestration layers. Systematic review of 141 multi-agent debate studies with a three-dimensional taxonomy over participants, interaction mechanisms, and agreement protocols, concluding the field has converged prematurely on one conventional design. Useful map for ... | Use Multi-Agent Debate Strategies: Survey, Taxonomy, and Challenges to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.26212; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,449 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1449 | 2026-07-30 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.26212 | [2607.26212] Multi-Agent Debate Strategies: Survey, Taxonomy, and Challenges | Multi-Agent Debate (MAD) is a promising paradigm for improving the accuracy and robustness of Large Language Model (LLM)-based agentic systems. It enables multiple agents to exchange arguments, critique each other's outputs, and iteratively converge towards a solution. However, research remains fragmented, with inconsi... | Quim Motger; Marc Oriol; Jordi Marco; Xavier Franch | 2026-07-28 | 2026 | arXiv | arXiv | Under review at ACM Computing Surveys | cs.SE | arxiv-api | 2607.26212 | 2026-09-08T16:27:48 | |||||||
ale-0750 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | Agent Manager | https://github.com/YoanWai/agent-manager | external | github.com | Terminal UI for operating a small agent fleet, surfaced on HN 2026-07-30. Each agent runs in an isolated tmux session that persists after the manager exits, and dead sessions revive with conversation history intact, so the fleet outlives the operator's terminal. | Terminal UI for operating a small agent fleet, surfaced on HN 2026-07-30. Each agent runs in an isolated tmux session that persists after the manager exits, and dead sessions revive with conversation history intact, so the fleet outlives the operator's terminal. | Terminal UI for operating a small agent fleet, surfaced on HN 2026-07-30. Each agent runs in an isolated tmux session that persists after the manager exits, and dead sessions revive with conversation history intact, so the fleet outlives the operator's terminal. | State persistence is explicit enough for repeated runs and handoff. Terminal UI for operating a small agent fleet, surfaced on HN 2026-07-30. Each agent runs in an isolated tmux session that persists after the manager exits, and dead sessions revive with conversation history intact, so the fleet outlives the operator's... | Use Agent Manager to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,450 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1450 | 2026-07-30 | Build | build | Choose runtimes, tools, and delegation surfaces. | state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/YoanWai/agent-manager | GitHub - YoanWai/agent-manager: The fastest workflow for every AI coding agent. Live status, quick prompts, worktrees, and diff review from one tmux TUI. · GitHub | The fastest workflow for every AI coding agent. Live status, quick prompts, worktrees, and diff review from one tmux TUI. - YoanWai/agent-manager | YoanWai/agent-manager | GitHub | html-meta | YoanWai/agent-manager | 2026-09-08T16:27:48 | ||||||||||||
ale-0751 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | TrueDeck | https://github.com/WutIsHummus/TrueDeck | external | github.com | Terminal-first workbench that puts Grok, Codex, Cursor, Claude, and Gemini in split panes on one codebase, each keeping its real TUI rather than being wrapped in a chat webview. | Terminal-first workbench that puts Grok, Codex, Cursor, Claude, and Gemini in split panes on one codebase, each keeping its real TUI rather than being wrapped in a chat webview. | Terminal-first workbench that puts Grok, Codex, Cursor, Claude, and Gemini in split panes on one codebase, each keeping its real TUI rather than being wrapped in a chat webview. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Terminal-first workbench that puts Grok, Codex, Cursor, Claude, and Gemini in split panes on one codebase, each keeping its real TUI rather than being wrapped in a chat webview. | Use TrueDeck to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,451 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1451 | 2026-07-30 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | source-implementation | A | ok | https://github.com/WutIsHummus/TrueDeck | GitHub - WutIsHummus/TrueDeck: Free multi-agent coding workbench: Grok/Codex/Claude/Cursor panes + TrueMemory (per-repo + global) · GitHub | Free multi-agent coding workbench: Grok/Codex/Claude/Cursor panes + TrueMemory (per-repo + global) - WutIsHummus/TrueDeck | WutIsHummus/TrueDeck | GitHub | html-meta | WutIsHummus/TrueDeck | 2026-09-08T16:27:48 | ||||||||||||
ale-0752 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | qm | https://github.com/yc-software/qm | external | github.com | Y Combinator's open-source agent platform for whole organizations rather than one operator. Each person and each Slack room gets its own scoped memory, files, keychain view, permissions, crons, web apps, and durable sandbox, so background work keeps running "while nobody's watching" without workspaces bleeding into eac... | Y Combinator's open-source agent platform for whole organizations rather than one operator. Each person and each Slack room gets its own scoped memory, files, keychain view, permissions, crons, web apps, and durable sandbox, so background work keeps running "while nobody's watching" without workspaces bleeding into eac... | Y Combinator's open-source agent platform for whole organizations rather than one operator. Each person and each Slack room gets its own scoped memory, files, keychain view, permissions, crons, web apps, and durable sandbox, so background work keeps running "while nobody's watching" without workspaces bleeding into eac... | Durable execution and replay are treated as first-class loop infrastructure. Y Combinator's open-source agent platform for whole organizations rather than one operator. Each person and each Slack room gets its own scoped memory, files, keychain view, permissions, crons, web apps, and durable sandbox, so background work... | Use qm to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,452 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1452 | 2026-08-02 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;context | builder | workflow | enabling | source-implementation | A | ok | https://github.com/yc-software/qm | GitHub - yc-software/qm: Multiplayer agent harness for work. · GitHub | Multiplayer agent harness for work. Contribute to yc-software/qm development by creating an account on GitHub. | yc-software/qm | GitHub | html-meta | yc-software/qm | 2026-09-08T16:27:48 | ||||||||||||
ale-0753 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | AgentRadio: Passive Awareness for Long-Horizon Multi-Agent Collaboration | https://arxiv.org/abs/2607.28430 | external | arxiv.org | Adds asynchronous message passing so coding agents keep background awareness of each other instead of exchanging information only at handoff boundaries. Four coordinated agents reach 62.1 percent on a long-horizon codebase-comprehension benchmark against 32.3 percent solo, and beat the same harness running a newer mode... | Adds asynchronous message passing so coding agents keep background awareness of each other instead of exchanging information only at handoff boundaries. Four coordinated agents reach 62.1 percent on a long-horizon codebase-comprehension benchmark against 32.3 percent solo, and beat the same harness running a newer mode... | Adds asynchronous message passing so coding agents keep background awareness of each other instead of exchanging information only at handoff boundaries. Four coordinated agents reach 62.1 percent on a long-horizon codebase-comprehension benchmark against 32.3 percent solo, and beat the same harness running a newer mode... | The work turns loop quality into a measurable task or score. Adds asynchronous message passing so coding agents keep background awareness of each other instead of exchanging information only at handoff boundaries. Four coordinated agents reach 62.1 percent on a long-horizon codebase-comprehension benchmark against 32.3... | Use AgentRadio: Passive Awareness for Long-Horizon Multi-Agent Collaboration to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.28430; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,453 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1453 | 2026-08-04 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;verification;escalation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.28430 | [2607.28430] AgentRadio: Passive Awareness for Long-Horizon Multi-Agent Collaboration | Understanding large codebases is a long-horizon task for Large Language Model (LLM) agents: answering a single question can require building and running the software, tracing execution across files, and synthesizing evidence over tens of minutes. On SWE-Atlas QnA, a benchmark of long-horizon questions over production r... | Xinxing Ren; Qianbo Zang; Ziyan Wang; Caelum Forder; Suman Deb; Peter Carroll; Zekun Guo | 2026-07-30 | 2026 | arXiv | arXiv | cs.MA | arxiv-api | 2607.28430 | 2026-09-08T16:27:48 | ||||||||
ale-0754 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | MANTA: Multi-Agent Network Topology Adaptation for Self-Evolving Multi-Agent Systems | https://arxiv.org/abs/2607.28527 | external | arxiv.org | Existing multi-agent systems treat communication topology as a fixed design choice or an offline optimization target. | Existing multi-agent systems treat communication topology as a fixed design choice or an offline optimization target. | Existing multi-agent systems treat communication topology as a fixed design choice or an offline optimization target. | The work separates roles across agents, verifiers, or orchestration layers. Existing multi-agent systems treat communication topology as a fixed design choice or an offline optimization target. | Use MANTA: Multi-Agent Network Topology Adaptation for Self-Evolving Multi-Agent Systems to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.28527; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,454 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1454 | 2026-08-05 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.28527 | [2607.28527] MANTA: Multi-Agent Network Topology Adaptation for Self-Evolving Multi-Agent Systems | Large language model-based multi-agent systems improve complex problem solving through task decomposition, agent specialization, information exchange, and intermediate validation. However, existing systems typically treat communication topology as a fixed design choice or an offline optimization target. We introduce MA... | Mao-xun Huang; Jerry Wang; Yi-Cheng Lai; Zhengxin Zhang; Claire Cardie; Hen-Hsen Huang | 2026-07-30 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2607.28527 | 2026-09-08T16:27:48 | ||||||||
ale-0755 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Tool | 🧰 | sol-advisor | https://github.com/DannyMac180/sol-advisor | external | github.com | Codex-native orchestration plugin that splits an agent run into an architect session that owns requirements, specs and acceptance, and bounded implementation lanes (native Terra lane, or an opt-in Luna lane using visible Codex app tasks) that only execute. | Codex-native orchestration plugin that splits an agent run into an architect session that owns requirements, specs and acceptance, and bounded implementation lanes (native Terra lane, or an opt-in Luna lane using visible Codex app tasks) that only execute. | Codex-native orchestration plugin that splits an agent run into an architect session that owns requirements, specs and acceptance, and bounded implementation lanes (native Terra lane, or an opt-in Luna lane using visible Codex app tasks) that only execute. | Orchestration and control flow are made explicit and inspectable. Codex-native orchestration plugin that splits an agent run into an architect session that owns requirements, specs and acceptance, and bounded implementation lanes (native Terra lane, or an opt-in Luna lane using visible Codex app tasks) that only execut... | Use sol-advisor to choose an implementation surface for repeatable agent work. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,455 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1455 | 2026-08-05 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | builder | workflow | enabling | source-implementation | A | ok | https://github.com/DannyMac180/sol-advisor | GitHub - DannyMac180/sol-advisor: Codex-native architect orchestration with Luna and Terra implementation lanes and mandatory fresh Sol review. · GitHub | Codex-native architect orchestration with Luna and Terra implementation lanes and mandatory fresh Sol review. - DannyMac180/sol-advisor | DannyMac180/sol-advisor | GitHub | html-meta | DannyMac180/sol-advisor | 2026-09-08T16:27:48 | ||||||||||||
ale-0756 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | SKIMIX: Multi-Agent Harness-Time Scaling with Skill Mixture for Dynamic Harness Engineering | https://arxiv.org/abs/2607.27994 | external | arxiv.org | Scales the harness rather than the model: embedding-based skill retrieval, submodular anti-dilution routing, and adaptive skill evolution let multiple agents refine over repeated rounds. Notably honest about limits, gains are large on open-ended math but flat or negative on multiple-choice, and mostly land in the first... | Scales the harness rather than the model: embedding-based skill retrieval, submodular anti-dilution routing, and adaptive skill evolution let multiple agents refine over repeated rounds. Notably honest about limits, gains are large on open-ended math but flat or negative on multiple-choice, and mostly land in the first... | Scales the harness rather than the model: embedding-based skill retrieval, submodular anti-dilution routing, and adaptive skill evolution let multiple agents refine over repeated rounds. Notably honest about limits, gains are large on open-ended math but flat or negative on multiple-choice, and mostly land in the first... | The work separates roles across agents, verifiers, or orchestration layers. Scales the harness rather than the model: embedding-based skill retrieval, submodular anti-dilution routing, and adaptive skill evolution let multiple agents refine over repeated rounds. Notably honest about limits, gains are large on open-ende... | Use SKIMIX: Multi-Agent Harness-Time Scaling with Skill Mixture for Dynamic Harness Engineering to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.27994; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,456 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1456 | 2026-08-05 | Build | build | Choose runtimes, tools, and delegation surfaces. | context;delegation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.27994 | [2607.27994] SKIMIX: Multi-Agent Harness-Time Scaling with Skill Mixture for Dynamic Harness Engineering | AI agents increasingly rely on large skill libraries, but selecting, combining, and maintaining skills remains difficult. We propose SKIMIX, a multi-agent framework in which agents with different skill portfolios collaborate through iterative refinement. SKIMIX combines embedding-based skill retrieval, submodular anti-... | Jia Luo | 2026-07-30 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2607.27994 | 2026-09-08T16:27:48 | ||||||||
ale-0757 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | An Empirical Study of Coordination Mode as the First-Class Citizen in From-Scratch Multi-Agent Coding | https://arxiv.org/abs/2607.27877 | external | arxiv.org | Introduces MSEval plus the LegoGent execution engine and holds the model fixed while varying 10 collaboration topologies, finding that organizational topology rivals model capability, score swings above 30 points. Structured pipelines converge fastest; heavy managerial layers hurt. | Introduces MSEval plus the LegoGent execution engine and holds the model fixed while varying 10 collaboration topologies, finding that organizational topology rivals model capability, score swings above 30 points. Structured pipelines converge fastest; heavy managerial layers hurt. | Introduces MSEval plus the LegoGent execution engine and holds the model fixed while varying 10 collaboration topologies, finding that organizational topology rivals model capability, score swings above 30 points. Structured pipelines converge fastest; heavy managerial layers hurt. | The work separates roles across agents, verifiers, or orchestration layers. Introduces MSEval plus the LegoGent execution engine and holds the model fixed while varying 10 collaboration topologies, finding that organizational topology rivals model capability, score swings above 30 points. Structured pipelines converge ... | Use An Empirical Study of Coordination Mode as the First-Class Citizen in From-Scratch Multi-Agent Coding to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.27877; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,457 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1457 | 2026-08-05 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.27877 | [2607.27877] An Empirical Study of Coordination Mode as the First-Class Citizen in From-Scratch Multi-Agent Coding | Multi-agent vibe coding promises to accelerate software development, yet existing benchmarks rely on synthetic environments that ignore practical time and monetary costs, conflate reasoning with communication, and reward only superficial completion. We introduce multi-agent from-scratch evaluation benchmark, MSEval, ev... | Yanyu Ren; Yunfeng Bai; Xizheng Wang; Li Chen; Dan Li | 2026-07-30 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2607.27877 | 2026-09-08T16:27:48 | ||||||||
ale-0758 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Scaling LLM-Driven Multi-Agent Systems: Design Principles and Architectural Scalability Analysis | https://arxiv.org/abs/2607.27942 | external | arxiv.org | Proposes four design principles, simplicity, elastic feedback, sequential workflows with optional loops, and summary-based communication, then tests them across four architecture configurations on standardized engineering tasks. | Proposes four design principles, simplicity, elastic feedback, sequential workflows with optional loops, and summary-based communication, then tests them across four architecture configurations on standardized engineering tasks. | Proposes four design principles, simplicity, elastic feedback, sequential workflows with optional loops, and summary-based communication, then tests them across four architecture configurations on standardized engineering tasks. | The work separates roles across agents, verifiers, or orchestration layers. Proposes four design principles, simplicity, elastic feedback, sequential workflows with optional loops, and summary-based communication, then tests them across four architecture configurations on standardized engineering tasks. | Use Scaling LLM-Driven Multi-Agent Systems: Design Principles and Architectural Scalability Analysis to choose an implementation surface for repeatable agent work. | Research source arXiv:2607.27942; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,458 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1458 | 2026-08-05 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;verification | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2607.27942 | [2607.27942] Scaling LLM-Driven Multi-Agent Systems: Design Principles and Architectural Scalability Analysis | LLM-based multi-agent systems have the potential to enable collective intelligence and scale toward solving highly complex tasks through coordinated ensembles of specialized agents. However, despite their theoretical potential, the architectural design space remains largely non-systematized and lacks broadly establishe... | Linus Sander; Fengjunjie Pan; Vahid Zolfaghari; Andre Schamschurko; Nenad Petrovic; Alois Knoll | 2026-07-30 | 2026 | arXiv | arXiv | cs.MA | arxiv-api | 2607.27942 | 2026-09-08T16:27:48 | ||||||||
ale-0759 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Scaling Managed Agents: Decoupling the Brain From the Hands | https://www.anthropic.com/engineering/managed-agents | external | www.anthropic.com | Virtualizes the agent into three interfaces: a stateless brain, tool sandboxes behind a single execute call, and a durable session event log held outside the harness. Because session state survives harness or container loss, each component becomes replaceable rather than a pet. | Virtualizes the agent into three interfaces: a stateless brain, tool sandboxes behind a single execute call, and a durable session event log held outside the harness. Because session state survives harness or container loss, each component becomes replaceable rather than a pet. | Virtualizes the agent into three interfaces: a stateless brain, tool sandboxes behind a single execute call, and a durable session event log held outside the harness. Because session state survives harness or container loss, each component becomes replaceable rather than a pet. | Durable execution and replay are treated as first-class loop infrastructure. Virtualizes the agent into three interfaces: a stateless brain, tool sandboxes behind a single execute call, and a durable session event log held outside the harness. Because session state survives harness or container loss, each component bec... | Use Scaling Managed Agents: Decoupling the Brain From the Hands to choose an implementation surface for repeatable agent work. | Contextual source from www.anthropic.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,459 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1459 | 2026-08-10 | Build | build | Choose runtimes, tools, and delegation surfaces. | workspace;state | builder | workflow | enabling | practitioner-analysis | B | ok | https://www.anthropic.com/engineering/managed-agents | Scaling Managed Agents: Decoupling the brain from the hands \ Anthropic | Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems. | Lance Martin; Gabe Cemaj; Michael Cohen | 2026-04-08 | 2026 | Anthropic Engineering | Anthropic | Author and date read from the post's own byline and dateline on anthropic.com/engineering; kept in code because automatic resolution falls back to the domain for this host. | publisher-meta | 2026-09-08T16:27:48 | |||||||||
ale-0760 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Blog | 📝 | Building a C Compiler With a Team of Parallel Claudes | https://www.anthropic.com/engineering/building-c-compiler | external | www.anthropic.com | Sixteen parallel agents built a 100k-line C compiler that compiles a bootable Linux kernel, coordinating through git rather than a scheduler: each claimed work by committing a file to a shared task directory. When all agents converged on one bug, GCC was used as an oracle to fan the work back out. | Sixteen parallel agents built a 100k-line C compiler that compiles a bootable Linux kernel, coordinating through git rather than a scheduler: each claimed work by committing a file to a shared task directory. When all agents converged on one bug, GCC was used as an oracle to fan the work back out. | Sixteen parallel agents built a 100k-line C compiler that compiles a bootable Linux kernel, coordinating through git rather than a scheduler: each claimed work by committing a file to a shared task directory. When all agents converged on one bug, GCC was used as an oracle to fan the work back out. | Shows how delegation, handoff, and workflow control turn one agent into a coordinated loop. Sixteen parallel agents built a 100k-line C compiler that compiles a bootable Linux kernel, coordinating through git rather than a scheduler: each claimed work by committing a file to a shared task directory. When all agents con... | Use Building a C Compiler With a Team of Parallel Claudes to choose an implementation surface for repeatable agent work. | Contextual source from www.anthropic.com; useful for practice signals or boundary conditions, not independent validation. | contextual | README.md | 1,460 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1460 | 2026-08-10 | Build | build | Choose runtimes, tools, and delegation surfaces. | delegation;state | builder | workflow | enabling | practitioner-analysis | B | ok | https://www.anthropic.com/engineering/building-c-compiler | Building a C compiler with a team of parallel Claudes \ Anthropic | Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems. | Nicholas Carlini | 2026-02-05 | 2026 | Anthropic Engineering | Anthropic | Author and date read from the post's own byline and dateline on anthropic.com/engineering; kept in code because automatic resolution falls back to the domain for this host. | publisher-meta | 2026-09-08T16:27:48 | |||||||||
ale-0761 | Orchestration And Multi-Agent Delegation | orchestration-and-multi-agent-delegation | Paper | 📄 | Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks | https://arxiv.org/abs/2608.05144 | external | arxiv.org | Manager, Planner, Engineer, and Reviewer run bounded missions over durable project state, with model weights fixed so self-evolution happens in runtime state and control policy rather than training. Memories, skills, verifiers, and routing decisions are admitted only after role-owned review, and rejected routes are ret... | Manager, Planner, Engineer, and Reviewer run bounded missions over durable project state, with model weights fixed so self-evolution happens in runtime state and control policy rather than training. Memories, skills, verifiers, and routing decisions are admitted only after role-owned review, and rejected routes are ret... | Manager, Planner, Engineer, and Reviewer run bounded missions over durable project state, with model weights fixed so self-evolution happens in runtime state and control policy rather than training. Memories, skills, verifiers, and routing decisions are admitted only after role-owned review, and rejected routes are ret... | Durable execution and replay are treated as first-class loop infrastructure. Manager, Planner, Engineer, and Reviewer run bounded missions over durable project state, with model weights fixed so self-evolution happens in runtime state and control policy rather than training. Memories, skills, verifiers, and routing dec... | Use Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks to choose an implementation surface for repeatable agent work. | Research source arXiv:2608.05144; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,461 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1461 | 2026-08-17 | Build | build | Choose runtimes, tools, and delegation surfaces. | context;state | researcher;evaluator | workflow | enabling | research-preprint | A | ok | https://arxiv.org/abs/2608.05144 | [2608.05144] Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks | Long-horizon reasoning requires an agentic runtime that can persist when evidence supports its current approach and pivot when measurements reveal failure, hidden constraints, or a misspecified objective. We present Argus, a persistent, self-evolving runtime in which Manager, Planner, Engineer, and Reviewer execute bou... | Boxiu Li; Zimo Wen; Yijia Fan; Chuan Wen; Fan Yang; Hangxi Guo; Jiaao Wu; Jiachen Zhang; Junxiang Lei; Mukai Li; Ruize Tang; Runjing Gu; Shibo Hu; Sihan Chen; Sufeng Guo; Wanbo Zhang; Xian Zhang; Xiaoyu Chen; Xuanhe Zhou; Xuyao Huang; Yifei Gao; Yifei Shen; Yilin Chen; Yuheng Wu; Yuzhe Zhang; Zelong Zhao; Zhijie Deng | 2026-08-05 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2608.05144 | 2026-09-08T16:27:48 | ||||||||
ale-0762 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SWE-bench | https://www.swebench.com/ | external | www.swebench.com | Benchmark for resolving real GitHub issues through code editing and tests. | Benchmark for resolving real GitHub issues through code editing and tests. | Benchmark for resolving real GitHub issues through code editing and tests. | The work turns loop quality into a measurable task or score. Benchmark for resolving real GitHub issues through code editing and tests. | Use SWE-bench to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,469 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1469 | Verify | verify | Gate progress with tests, evals, and evidence. | intake;verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://www.swebench.com/ | SWE-bench Leaderboards | swebench.com | domain-fallback | 2026-09-08T16:27:48 | ||||||||||||||||
ale-0763 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? | https://arxiv.org/abs/2310.06770 | external | arxiv.org | Original SWE-bench paper. | Original SWE-bench paper. | Original SWE-bench paper. | Links loop design to measurable tasks where progress and failure can be compared. Original SWE-bench paper. | Use SWE-bench: Can Language Models Resolve Real-World GitHub Issues? to measure progress and gate completion with repeatable evidence. | Research source arXiv:2310.06770; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,470 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1470 | Verify | verify | Gate progress with tests, evals, and evidence. | intake | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://proceedings.iclr.cc/paper_files/paper/2024/hash/edac78c3e300629acfe6cbe9ca88fb84-Abstract-Conference.html | [2310.06770] SWE-bench: Can Language Models Resolve Real-World GitHub Issues? | Language models have outpaced our ability to evaluate them effectively, but for their future development it is essential to study the frontier of their capabilities. We find real-world software engineering to be a rich, sustainable, and challenging testbed for evaluating the next generation of language models. To this ... | Carlos E. Jimenez; John Yang; Alexander Wettig; Shunyu Yao; Kexin Pei; Ofir Press; Karthik Narasimhan | 2024 | 2024 | International Conference on Learning Representations (ICLR) | International Conference on Learning Representations | Published in International Conference on Learning Representations (ICLR); the linked arXiv record remains available for open access. | cs.CL | ICLR proceedings record | 2310.06770 | 2026-09-08T16:27:48 | ||||||||
ale-0764 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | SWE-bench Goes Live | https://arxiv.org/abs/2505.23419 | external | arxiv.org | Dynamic benchmark designed to reduce overfitting to static issue sets. | Dynamic benchmark designed to reduce overfitting to static issue sets. | Dynamic benchmark designed to reduce overfitting to static issue sets. | The work turns loop quality into a measurable task or score. Dynamic benchmark designed to reduce overfitting to static issue sets. | Use SWE-bench Goes Live to measure progress and gate completion with repeatable evidence. | Research source arXiv:2505.23419; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,471 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1471 | Verify | verify | Gate progress with tests, evals, and evidence. | intake;verification | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://proceedings.neurips.cc/paper_files/paper/2025/hash/d83c4a745789690f82e86d0ef752ae7c-Abstract-Datasets_and_Benchmarks_Track.html | [2505.23419] SWE-bench Goes Live! | The issue-resolving task, where a model generates patches to fix real-world bugs, has emerged as a critical benchmark for evaluating the capabilities of large language models (LLMs). While SWE-bench and its variants have become standard in this domain, they suffer from key limitations: they have not been updated since ... | Linghao Zhang; Shilin He; Chaoyun Zhang; Yu Kang; Bowen Li; Chengxing Xie; Junhao Wang; Maoquan Wang; Yufan Huang; Shengyu Fu; Elsie Nallipogu; Qingwei Lin; Yingnong Dang; Saravan Rajmohan; Dongmei Zhang | 2025 | 2025 | Advances in Neural Information Processing Systems 38: Datasets and Benchmarks Track (NeurIPS) | Neural Information Processing Systems Foundation | Published in Advances in Neural Information Processing Systems 38: Datasets and Benchmarks Track (NeurIPS); the linked arXiv record remains available for open access. | cs.SE | NeurIPS proceedings record | 2505.23419 | 2026-09-08T16:27:48 | ||||||||
ale-0765 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | Terminal-Bench | https://www.tbench.ai/ | external | www.tbench.ai | Benchmark for agents operating in terminal environments. | Benchmark for agents operating in terminal environments. | Benchmark for agents operating in terminal environments. | The work turns loop quality into a measurable task or score. Benchmark for agents operating in terminal environments. | Use Terminal-Bench to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,472 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1472 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://www.tbench.ai/ | TERMINAL-BENCH | A benchmark to measure and evolve with the frontier of agent work | TERMINAL-BENCH | html-meta | 2026-09-08T16:27:48 | |||||||||||||||
ale-0766 | Benchmarks And Evaluation | benchmarks-and-evaluation | Tool | 🧰 | Terminal-Bench repository | https://github.com/harbor-framework/terminal-bench | external | github.com | Open-source benchmark and harness for hard terminal tasks. | Open-source benchmark and harness for hard terminal tasks. | Open-source benchmark and harness for hard terminal tasks. | The work turns loop quality into a measurable task or score. Open-source benchmark and harness for hard terminal tasks. | Use Terminal-Bench repository to measure progress and gate completion with repeatable evidence. | Inspectable GitHub source; popularity is context, not proof of reliability. | medium | README.md | 1,473 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1473 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | builder;evaluator | evaluation | enabling | source-implementation | A | ok | https://github.com/harbor-framework/terminal-bench | GitHub - harbor-framework/terminal-bench: Measuring and evolving with the frontier of agent work · GitHub | Measuring and evolving with the frontier of agent work - harbor-framework/terminal-bench | harbor-framework/terminal-bench | GitHub | html-meta | harbor-framework/terminal-bench | 2026-09-08T16:27:48 | |||||||||||||
ale-0767 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | AgentBench | https://arxiv.org/abs/2308.03688 | external | arxiv.org | Multi-environment benchmark for evaluating LLMs as agents. | Multi-environment benchmark for evaluating LLMs as agents. | Multi-environment benchmark for evaluating LLMs as agents. | The work turns loop quality into a measurable task or score. Multi-environment benchmark for evaluating LLMs as agents. | Use AgentBench to measure progress and gate completion with repeatable evidence. | Research source arXiv:2308.03688; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,474 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1474 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://proceedings.iclr.cc/paper_files/paper/2024/hash/e9df36b21ff4ee211a8b71ee8b7e9f57-Abstract-Conference.html | [2308.03688] AgentBench: Evaluating LLMs as Agents | The potential of Large Language Model (LLM) as agents has been widely acknowledged recently. Thus, there is an urgent need to quantitatively \textit{evaluate LLMs as agents} on challenging tasks in interactive environments. We present AgentBench, a multi-dimensional benchmark that consists of 8 distinct environments to... | Xiao Liu; Hao Yu; Hanchen Zhang; Yifan Xu; Xuanyu Lei; Hanyu Lai; Yu Gu; Hangliang Ding; Kaiwen Men; Kejuan Yang; Shudan Zhang; Xiang Deng; Aohan Zeng; Zhengxiao Du; Chenhui Zhang; Sheng Shen; Tianjun Zhang; Yu Su; Huan Sun; Minlie Huang; Yuxiao Dong; Jie Tang | 2024 | 2024 | International Conference on Learning Representations (ICLR) | International Conference on Learning Representations | Published in International Conference on Learning Representations (ICLR); the linked arXiv record remains available for open access. | cs.AI | ICLR proceedings record | 2308.03688 | 2026-09-08T16:27:48 | ||||||||
ale-0768 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | WebArena | https://arxiv.org/abs/2307.13854 | external | arxiv.org | Realistic web environment for autonomous agents. | Realistic web environment for autonomous agents. | Realistic web environment for autonomous agents. | Links loop design to measurable tasks where progress and failure can be compared. Realistic web environment for autonomous agents. | Use WebArena to measure progress and gate completion with repeatable evidence. | Research source arXiv:2307.13854; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,475 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1475 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://proceedings.iclr.cc/paper_files/paper/2024/hash/4410c0711e9154a7a2d26f9b3816d1ef-Abstract-Conference.html | [2307.13854] WebArena: A Realistic Web Environment for Building Autonomous Agents | With advances in generative AI, there is now potential for autonomous agents to manage daily tasks via natural language commands. However, current agents are primarily created and tested in simplified synthetic environments, leading to a disconnect with real-world scenarios. In this paper, we build an environment for l... | Shuyan Zhou; Frank F. Xu; Hao Zhu; Xuhui Zhou; Robert Lo; Abishek Sridhar; Xianyi Cheng; Tianyue Ou; Yonatan Bisk; Daniel Fried; Uri Alon; Graham Neubig | 2024 | 2024 | International Conference on Learning Representations (ICLR) | International Conference on Learning Representations | Published in International Conference on Learning Representations (ICLR); the linked arXiv record remains available for open access. | cs.AI | ICLR proceedings record | 2307.13854 | 2026-09-08T16:27:48 | ||||||||
ale-0769 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | OSWorld | https://arxiv.org/abs/2404.07972 | external | arxiv.org | Benchmark for multimodal agents operating full computer environments. | Benchmark for multimodal agents operating full computer environments. | Benchmark for multimodal agents operating full computer environments. | The work turns loop quality into a measurable task or score. Benchmark for multimodal agents operating full computer environments. | Use OSWorld to measure progress and gate completion with repeatable evidence. | Research source arXiv:2404.07972; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,476 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1476 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://proceedings.neurips.cc/paper_files/paper/2024/hash/5d413e48f84dc61244b6be550f1cd8f5-Abstract-Datasets_and_Benchmarks_Track.html | [2404.07972] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments | Autonomous agents that accomplish complex computer tasks with minimal human interventions have the potential to transform human-computer interaction, significantly enhancing accessibility and productivity. However, existing benchmarks either lack an interactive environment or are limited to environments specific to cer... | Tianbao Xie; Danyang Zhang; Jixuan Chen; Xiaochuan Li; Siheng Zhao; Ruisheng Cao; Toh Jing Hua; Zhoujun Cheng; Dongchan Shin; Fangyu Lei; Yitao Liu; Yiheng Xu; Shuyan Zhou; Silvio Savarese; Caiming Xiong; Victor Zhong; Tao Yu | 2024 | 2024 | Advances in Neural Information Processing Systems 37: Datasets and Benchmarks Track (NeurIPS) | Neural Information Processing Systems Foundation | 10.52202/079017-1650 | Published in Advances in Neural Information Processing Systems 37: Datasets and Benchmarks Track (NeurIPS); the linked arXiv record remains available for open access. | cs.AI | NeurIPS proceedings and DOI records | 2404.07972 | 2026-09-08T16:27:48 | |||||||
ale-0770 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | ToolBench | https://arxiv.org/abs/2307.16789 | external | arxiv.org | Tool-use benchmark and dataset for tool-augmented agents. | Tool-use benchmark and dataset for tool-augmented agents. | Tool-use benchmark and dataset for tool-augmented agents. | Packages the evidence as queryable CSV and JSONL rather than only a rendered page. Tool-use benchmark and dataset for tool-augmented agents. | Use ToolBench to measure progress and gate completion with repeatable evidence. | Research source arXiv:2307.16789; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,477 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1477 | Verify | verify | Gate progress with tests, evals, and evidence. | workspace;verification | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://proceedings.iclr.cc/paper_files/paper/2024/hash/28e50ee5b72e90b50e7196fde8ea260e-Abstract-Conference.html | [2307.16789] ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs | Despite the advancements of open-source large language models (LLMs), e.g., LLaMA, they remain significantly limited in tool-use capabilities, i.e., using external tools (APIs) to fulfill human instructions. The reason is that current instruction tuning largely focuses on basic language tasks but ignores the tool-use d... | Yujia Qin; Shihao Liang; Yining Ye; Kunlun Zhu; Lan Yan; Yaxi Lu; Yankai Lin; Xin Cong; Xiangru Tang; Bill Qian; Sihan Zhao; Lauren Hong; Runchu Tian; Ruobing Xie; Jie Zhou; Mark Gerstein; Dahai Li; Zhiyuan Liu; Maosong Sun | 2024 | 2024 | International Conference on Learning Representations (ICLR) | International Conference on Learning Representations | Published in International Conference on Learning Representations (ICLR); the linked arXiv record remains available for open access. | cs.AI | ICLR proceedings record | 2307.16789 | 2026-09-08T16:27:48 | ||||||||
ale-0771 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | GAIA | https://arxiv.org/abs/2311.12983 | external | arxiv.org | Benchmark for general AI assistants requiring reasoning, tool use, and multi-step work. | Benchmark for general AI assistants requiring reasoning, tool use, and multi-step work. | Benchmark for general AI assistants requiring reasoning, tool use, and multi-step work. | The work turns loop quality into a measurable task or score. Benchmark for general AI assistants requiring reasoning, tool use, and multi-step work. | Use GAIA to measure progress and gate completion with repeatable evidence. | Research source arXiv:2311.12983; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,478 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1478 | Verify | verify | Gate progress with tests, evals, and evidence. | workspace;verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2311.12983 | [2311.12983] GAIA: a benchmark for General AI Assistants | We introduce GAIA, a benchmark for General AI Assistants that, if solved, would represent a milestone in AI research. GAIA proposes real-world questions that require a set of fundamental abilities such as reasoning, multi-modality handling, web browsing, and generally tool-use proficiency. GAIA questions are conceptual... | Grégoire Mialon; Clémentine Fourrier; Craig Swift; Thomas Wolf; Yann LeCun; Thomas Scialom | 2023-11-21 | 2023 | arXiv | arXiv | cs.CL | arxiv-api | 2311.12983 | 2026-09-08T16:27:48 | |||||||||
ale-0772 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | Tau-bench | https://arxiv.org/abs/2406.12045 | external | arxiv.org | Benchmark for tool-agent-user interactions in realistic domains. | Benchmark for tool-agent-user interactions in realistic domains. | Benchmark for tool-agent-user interactions in realistic domains. | The work turns loop quality into a measurable task or score. Benchmark for tool-agent-user interactions in realistic domains. | Use Tau-bench to measure progress and gate completion with repeatable evidence. | Research source arXiv:2406.12045; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,479 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1479 | Verify | verify | Gate progress with tests, evals, and evidence. | workspace;verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2406.12045 | [2406.12045] $τ$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains | Existing benchmarks do not test language agents on their interaction with human users or ability to follow domain-specific rules, both of which are vital for deploying them in real world applications. We propose $\tau$-bench, a benchmark emulating dynamic conversations between a user (simulated by language models) and ... | Shunyu Yao; Noah Shinn; Pedram Razavi; Karthik Narasimhan | 2024-06-17 | 2024 | arXiv | arXiv | cs.AI | arxiv-api | 2406.12045 | 2026-09-08T16:27:48 | |||||||||
ale-0773 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | VisualWebArena | https://arxiv.org/abs/2401.13649 | external | arxiv.org | Visually grounded web-agent benchmark extending WebArena. | Visually grounded web-agent benchmark extending WebArena. | Visually grounded web-agent benchmark extending WebArena. | The work turns loop quality into a measurable task or score. Visually grounded web-agent benchmark extending WebArena. | Use VisualWebArena to measure progress and gate completion with repeatable evidence. | Research source arXiv:2401.13649; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,480 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1480 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://aclanthology.org/2024.acl-long.50/ | [2401.13649] VisualWebArena: Evaluating Multimodal Agents on Realistic Visual Web Tasks | Autonomous agents capable of planning, reasoning, and executing actions on the web offer a promising avenue for automating computer tasks. However, the majority of existing benchmarks primarily focus on text-based agents, neglecting many natural tasks that require visual information to effectively solve. Given that mos... | Jing Yu Koh; Robert Lo; Lawrence Jang; Vikram Duvvur; Ming Chong Lim; Po-Yu Huang; Graham Neubig; Shuyan Zhou; Ruslan Salakhutdinov; Daniel Fried | 2024 | 2024 | Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL) | Association for Computational Linguistics | 10.18653/v1/2024.acl-long.50 | Published in Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL); the linked arXiv record remains available for open access. | cs.LG | ACL Anthology and DOI records | 2401.13649 | 2026-09-08T16:27:48 | |||||||
ale-0774 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | AppWorld | https://arxiv.org/abs/2407.18901 | external | arxiv.org | Benchmark of interactive app tasks with state-based and execution-based evaluation. | Benchmark of interactive app tasks with state-based and execution-based evaluation. | Benchmark of interactive app tasks with state-based and execution-based evaluation. | Evaluation data is used as the feedback signal for improving loop behavior. Benchmark of interactive app tasks with state-based and execution-based evaluation. | Use AppWorld to measure progress and gate completion with repeatable evidence. | Research source arXiv:2407.18901; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,481 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1481 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;state | researcher;evaluator | evaluation | enabling | research-paper | A | ok | https://aclanthology.org/2024.acl-long.850/ | [2407.18901] AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agents | Autonomous agents that address day-to-day digital tasks (e.g., ordering groceries for a household), must not only operate multiple apps (e.g., notes, messaging, shopping app) via APIs, but also generate rich code with complex control flow in an iterative manner based on their interaction with the environment. However, ... | Harsh Trivedi; Tushar Khot; Mareike Hartmann; Ruskin Manku; Vinty Dong; Edward Li; Shashank Gupta; Ashish Sabharwal; Niranjan Balasubramanian | 2024 | 2024 | Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL) | Association for Computational Linguistics | 10.18653/v1/2024.acl-long.850 | Published in Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL); the linked arXiv record remains available for open access. | cs.SE | ACL Anthology and DOI records | 2407.18901 | 2026-09-08T16:27:48 | |||||||
ale-0775 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | Vending-Bench | https://arxiv.org/abs/2502.15840 | external | arxiv.org | Benchmark for long-term coherence of autonomous agents; documents how small errors compound over very long loop horizons. | Benchmark for long-term coherence of autonomous agents; documents how small errors compound over very long loop horizons. | Benchmark for long-term coherence of autonomous agents; documents how small errors compound over very long loop horizons. | The work turns loop quality into a measurable task or score. Benchmark for long-term coherence of autonomous agents; documents how small errors compound over very long loop horizons. | Use Vending-Bench to measure progress and gate completion with repeatable evidence. | Research source arXiv:2502.15840; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,482 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1482 | Verify | verify | Gate progress with tests, evals, and evidence. | context;verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2502.15840 | [2502.15840] Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents | While Large Language Models (LLMs) can exhibit impressive proficiency in isolated, short-term tasks, they often fail to maintain coherent performance over longer time horizons. In this paper, we present Vending-Bench, a simulated environment designed to specifically test an LLM-based agent's ability to manage a straigh... | Axel Backlund; Lukas Petersson | 2025-02-20 | 2025 | arXiv | arXiv | cs.AI | arxiv-api | 2502.15840 | 2026-09-08T16:27:48 | |||||||||
ale-0776 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | Vending-Bench leaderboard | https://andonlabs.com/evals/vending-bench | external | andonlabs.com | Live long-horizon coherence results from Andon Labs. | Live long-horizon coherence results from Andon Labs. | Live long-horizon coherence results from Andon Labs. | The work turns loop quality into a measurable task or score. Live long-horizon coherence results from Andon Labs. | Use Vending-Bench leaderboard to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,483 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1483 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://andonlabs.com/evals/vending-bench | Vending-Bench: Testing long-term coherence in agents | Andon Labs | How do agents act over very long horizons? We answer this by letting agents manage a simulated vending machine business. The agents need to handle ordering, inventory management, and pricing over long context horizons to successfully make money. | andonlabs.com | domain-fallback | 2026-09-08T16:27:48 | |||||||||||||||
ale-0777 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarios | https://arxiv.org/abs/2512.18470 | external | arxiv.org | Release-note-derived evolution tasks where agents score far below isolated-issue benchmarks, quantifying the long-horizon gap loops must manage. | Release-note-derived evolution tasks where agents score far below isolated-issue benchmarks, quantifying the long-horizon gap loops must manage. | Release-note-derived evolution tasks where agents score far below isolated-issue benchmarks, quantifying the long-horizon gap loops must manage. | The work turns loop quality into a measurable task or score. Release-note-derived evolution tasks where agents score far below isolated-issue benchmarks, quantifying the long-horizon gap loops must manage. | Use SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarios to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,484 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1484 | Verify | verify | Gate progress with tests, evals, and evidence. | intake;verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2512.18470 | [2512.18470] SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarios | Existing benchmarks for AI coding agents focus on isolated, single-issue tasks such as fixing a bug or adding a small feature. However, real-world software engineering is a long-horizon endeavor: developers interpret high-level requirements, coordinate changes across many files, and evolve codebases over multiple itera... | Tue Le; Minh V. T. Thai; Dung Nguyen Manh; Huy Phan Nhat; Nghi D. Q. Bui | 2025-12-20 | 2025 | arXiv | arXiv | cs.SE | arxiv-api | 2512.18470 | 2026-09-08T16:27:48 | |||||||||
ale-0778 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | EvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification | https://arxiv.org/abs/2604.01687 | external | arxiv.org | A skill generator and a co-evolving surrogate verifier improve multi-file skill packages over iterations, evaluated on the SkillsBench benchmark of structured skill bundles. | A skill generator and a co-evolving surrogate verifier improve multi-file skill packages over iterations, evaluated on the SkillsBench benchmark of structured skill bundles. | A skill generator and a co-evolving surrogate verifier improve multi-file skill packages over iterations, evaluated on the SkillsBench benchmark of structured skill bundles. | Verification is promoted from a final check to a loop-control signal. A skill generator and a co-evolving surrogate verifier improve multi-file skill packages over iterations, evaluated on the SkillsBench benchmark of structured skill bundles. | Use EvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification to measure progress and gate completion with repeatable evidence. | Research source arXiv:2604.01687; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,485 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1485 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2604.01687 | [2604.01687] CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification | Anthropic proposes the concept of skills for LLM agents to tackle multi-step professional tasks that simple tool invocations cannot address. A tool is a single, self-contained function, whereas a skill is a structured bundle of interdependent multi-file artifacts. Currently, skill generation is not only label-intensive... | Hanrong Zhang; Shicheng Fan; Henry Peng Zou; Yankai Chen; Zhenting Wang; Jiayu Zhou; Chengze Li; Wei-Chieh Huang; Yifei Yao; Kening Zheng; Xue; Liu; Xiaoxiao Li; Philip S. Yu | 2026-04-02 | 2026 | arXiv | arXiv | COLM accepted | cs.AI | arxiv-api | 2604.01687 | 2026-09-08T16:27:48 | ||||||||
ale-0779 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SaaSBench: Coding Agents in Long-Horizon Enterprise SaaS Engineering | https://arxiv.org/abs/2605.17526 | external | arxiv.org | Benchmark for agents on multi-dependency, interactive enterprise tasks, with automated evaluation that probes where long-horizon loops break down. | Benchmark for agents on multi-dependency, interactive enterprise tasks, with automated evaluation that probes where long-horizon loops break down. | Benchmark for agents on multi-dependency, interactive enterprise tasks, with automated evaluation that probes where long-horizon loops break down. | Evaluation data is used as the feedback signal for improving loop behavior. Benchmark for agents on multi-dependency, interactive enterprise tasks, with automated evaluation that probes where long-horizon loops break down. | Use SaaSBench: Coding Agents in Long-Horizon Enterprise SaaS Engineering to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,486 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1486 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2605.17526 | [2605.17526] SaaSBench: Exploring the Boundaries of Coding Agents in Long-Horizon Enterprise SaaS Engineering | As autonomous coding agents become capable of handling increasingly long-horizon tasks, they have gradually demonstrated the potential to complete end-to-end software development. Although existing benchmarks have recently evolved from localized code editing to from-scratch project generation, they remain confined to s... | Qingnan Ren; Shun Zou; Shiting Huang; Ziao Zhang; Kou Shi; Zhen Fang; Yiming Zhao; Yu Zeng; Qisheng Su; Lin Chen; Yong Wang; Zehui Chen; Xiangxiang Chu; Feng Zhao | 2026-05-17 | 2026 | arXiv | arXiv | cs.SE | arxiv-api | 2605.17526 | 2026-09-08T16:27:48 | |||||||||
ale-0780 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | RoadmapBench: Evaluating Long-Horizon Agentic Software Development Across Version Upgrades | https://arxiv.org/abs/2605.15846 | external | arxiv.org | 115 real version-upgrade tasks across 17 repositories requiring multi-file changes (median ~3,700 lines), stressing how far agent loops sustain coherent, large-scale work. | 115 real version-upgrade tasks across 17 repositories requiring multi-file changes (median ~3,700 lines), stressing how far agent loops sustain coherent, large-scale work. | 115 real version-upgrade tasks across 17 repositories requiring multi-file changes (median ~3,700 lines), stressing how far agent loops sustain coherent, large-scale work. | The work targets tasks that exceed a single context window or prompt session. 115 real version-upgrade tasks across 17 repositories requiring multi-file changes (median ~3,700 lines), stressing how far agent loops sustain coherent, large-scale work. | Use RoadmapBench: Evaluating Long-Horizon Agentic Software Development Across Version Upgrades to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,487 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1487 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2605.15846 | [2605.15846] RoadmapBench: Evaluating Long-Horizon Agentic Software Development Across Version Upgrades | Coding agents are increasingly deployed in real software development, where a single version iteration requires months of coordinated work across many files. However, most existing benchmarks focus predominantly on single-issue bug fixes from Python repositories, with coarse pass/fail evaluation outcomes, and thus fail... | Xinbo Xu; Ruihan Yang; Haiyang Shen; Wendong Xu; Bofei Gao; Ruoyu Wu; Kean Shi; Weichu Xie; Xuanzhong Chen; Ming Wu; Jason Zeng; Michael Heinrich; Elvis Zhang; Liang Chen; Kuan Li; Baobao Chang | 2026-05-15 | 2026 | arXiv | arXiv | 30 pages, 15 figures | cs.SE | arxiv-api | 2605.15846 | 2026-09-08T16:27:48 | ||||||||
ale-0781 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | RefactorBench: Evaluating Stateful Reasoning in Language Agents Through Code | https://arxiv.org/abs/2503.07832 | external | arxiv.org | Multi-file refactoring tasks that require tracking and carrying state across many steps, isolating the durable-state weakness that breaks long agent loops. | Multi-file refactoring tasks that require tracking and carrying state across many steps, isolating the durable-state weakness that breaks long agent loops. | Multi-file refactoring tasks that require tracking and carrying state across many steps, isolating the durable-state weakness that breaks long agent loops. | Durable execution and replay are treated as first-class loop infrastructure. Multi-file refactoring tasks that require tracking and carrying state across many steps, isolating the durable-state weakness that breaks long agent loops. | Use RefactorBench: Evaluating Stateful Reasoning in Language Agents Through Code to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,488 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1488 | Verify | verify | Gate progress with tests, evals, and evidence. | state | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://proceedings.iclr.cc/paper_files/paper/2025/hash/6b44ee74539ea77d6a0d50d468724371-Abstract-Conference.html | [2503.07832] RefactorBench: Evaluating Stateful Reasoning in Language Agents Through Code | Recent advances in language model (LM) agents and function calling have enabled autonomous, feedback-driven systems to solve problems across various digital domains. To better understand the unique limitations of LM agents, we introduce RefactorBench, a benchmark consisting of 100 large handcrafted multi-file refactori... | Dhruv Gautam; Spandan Garg; Jinu Jang; Neel Sundaresan; Roshanak Zilouchian Moghaddam | 2025 | 2025 | International Conference on Learning Representations (ICLR) | International Conference on Learning Representations | Published in International Conference on Learning Representations (ICLR); the linked arXiv record remains available for open access. | cs.AI | ICLR proceedings record | 2503.07832 | 2026-09-08T16:27:48 | ||||||||
ale-0782 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents | https://arxiv.org/abs/2606.22678 | external | arxiv.org | Scores planning, verification coverage, recovery, abstention, and atomic transitions (not just whether code passes), measuring the loop discipline that separates reliable agents from reckless trial-and-error. | Scores planning, verification coverage, recovery, abstention, and atomic transitions (not just whether code passes), measuring the loop discipline that separates reliable agents from reckless trial-and-error. | Scores planning, verification coverage, recovery, abstention, and atomic transitions (not just whether code passes), measuring the loop discipline that separates reliable agents from reckless trial-and-error. | Verification is promoted from a final check to a loop-control signal. Scores planning, verification coverage, recovery, abstention, and atomic transitions (not just whether code passes), measuring the loop discipline that separates reliable agents from reckless trial-and-error. | Use RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,489 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1489 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2606.22678 | [2606.22678] RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents | Agentic coding harnesses - such as Agent-Skills, Superpowers, and Agent-Rigor - are increasingly deployed to augment underlying LLMs for real-world software engineering tasks. Existing benchmarks evaluate these agents almost exclusively on outcome correctness: whether generated code passes tests or resolves issues. We ... | Meher Bhaskar Madiraju; Meher Sai Preetam Madiraju | 2026-06-21 | 2026 | arXiv | arXiv | 9 pages, 7 tables, 1 figure | cs.SE | arxiv-api | 2606.22678 | 2026-09-08T16:27:48 | ||||||||
ale-0783 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks | https://arxiv.org/abs/2603.24755 | external | arxiv.org | Quantifies structural erosion and verbosity creep across iteration checkpoints in native harnesses like Claude Code and Codex, evidence for why loops need verification and budgets. | Quantifies structural erosion and verbosity creep across iteration checkpoints in native harnesses like Claude Code and Codex, evidence for why loops need verification and budgets. | Quantifies structural erosion and verbosity creep across iteration checkpoints in native harnesses like Claude Code and Codex, evidence for why loops need verification and budgets. | Checkpointed state makes long-running agent work recoverable across failures. Quantifies structural erosion and verbosity creep across iteration checkpoints in native harnesses like Claude Code and Codex, evidence for why loops need verification and budgets. | Use SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,490 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1490 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;state;budget | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2603.24755 | [2603.24755] SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks | Software development is iterative, yet agentic coding benchmarks hide design issues through their single-shot setup. Recent iterative benchmarks attempt to remedy this but heavily constrain an agent's design decision space, making it impossible to faithfully measure how their decisions shape future extensions. We intro... | Gabriel Orlanski; Devjeet Roy; Alexander Yun; Changho Shin; Alex Gu; Albert Ge; Dyah Adila; Nicholas Roberts; Frederic Sala; Aws Albarghouthi | 2026-03-25 | 2026 | arXiv | arXiv | 10.5281/zenodo.18405900, | Code and Leaderboards are located at https://www.scbench.ai | cs.SE | arxiv-api | 2603.24755 | 2026-09-08T16:27:48 | |||||||
ale-0784 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | LongCLI-Bench: A Preliminary Benchmark for Long-horizon Agentic Programming in Command-Line Interfaces | https://arxiv.org/abs/2602.14337 | external | arxiv.org | Long-horizon CLI tasks where most runs stall below 30% completion, mapping where unattended loops break down. | Long-horizon CLI tasks where most runs stall below 30% completion, mapping where unattended loops break down. | Long-horizon CLI tasks where most runs stall below 30% completion, mapping where unattended loops break down. | The work turns loop quality into a measurable task or score. Long-horizon CLI tasks where most runs stall below 30% completion, mapping where unattended loops break down. | Use LongCLI-Bench: A Preliminary Benchmark for Long-horizon Agentic Programming in Command-Line Interfaces to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,491 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1491 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;exit | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://aclanthology.org/2026.findings-acl.1497/ | [2602.14337] LongCLI-Bench: A Preliminary Benchmark and Study for Long-horizon Agentic Programming in Command-Line Interfaces | Recent advances in AI-assisted programming have empowered agents to execute complex workflows via command-line interfaces, however, existing benchmarks are limited by short task horizons, data contamination from GitHub scraping, and a lack of fine-grained evaluation metrics, fail to rigorously evaluate the long-horizon... | Yukang Feng; Jianwen Sun; Zelai Yang; Jiaxin Ai; Chuanhao Li; Zizhen Li; Fanrui Zhang; Kang He; Rui Ma; Jifan Lin; Jie Sun; Yang Xiao; Sizhuo Zhou; Wenxiao Wu; Yiming Liu; Pengfei Liu; Yu Qiao; Shenglin Zhang; Kaipeng Zhang | 2026 | 2026 | Findings of the Association for Computational Linguistics: ACL | Association for Computational Linguistics | 10.18653/v1/2026.findings-acl.1497 | Published in Findings of the Association for Computational Linguistics: ACL; the linked arXiv record remains available for open access. | cs.SE | ACL Anthology and DOI records | 2602.14337 | 2026-09-08T16:27:48 | |||||||
ale-0785 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | Can LLM-as-a-Judge Reliably Verify Rubrics in Agentic Scenarios? | https://arxiv.org/abs/2606.29920 | external | arxiv.org | Benchmark of 2,458 instances across research and coding domains measuring how reliably LLM judges verify rubrics on agent outputs, finding substantial noise even in strong models and quantifying the trade-offs of prompt design, batched evaluation, and majority voting. | Benchmark of 2,458 instances across research and coding domains measuring how reliably LLM judges verify rubrics on agent outputs, finding substantial noise even in strong models and quantifying the trade-offs of prompt design, batched evaluation, and majority voting. | Benchmark of 2,458 instances across research and coding domains measuring how reliably LLM judges verify rubrics on agent outputs, finding substantial noise even in strong models and quantifying the trade-offs of prompt design, batched evaluation, and majority voting. | Evaluation data is used as the feedback signal for improving loop behavior. Benchmark of 2,458 instances across research and coding domains measuring how reliably LLM judges verify rubrics on agent outputs, finding substantial noise even in strong models and quantifying the trade-offs of prompt design, batched evaluati... | Use Can LLM-as-a-Judge Reliably Verify Rubrics in Agentic Scenarios? to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,492 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1492 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2606.29920 | [2606.29920] Can LLM-as-a-Judge Reliably Verify Rubrics in Agentic Scenarios? | Rubric-based scoring has become a widely used paradigm in model evaluation, typically with LLM-as-a-Judge (LaaJ) for rubric scoring. However, the reliability of LaaJ for rubric scoring remains underexplored. This concern is especially pronounced in agentic scenarios, where long, complex outputs further challenge reliab... | Yangda Peng; Yunjia Qi; Haotian Xia; Guanzhong He; Xintong Shi; Richeng Xuan; Songyuanyi Lu; Yixian Liu; Zhichao Hu; Yuhong Liu; Hao Peng | 2026-06-29 | 2026 | arXiv | arXiv | Accepted to EMNLP 2026 | cs.CL | arxiv-api | 2606.29920 | 2026-09-08T16:27:48 | ||||||||
ale-0786 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SentinelBench: A Benchmark for Long-Running Monitoring Agents | https://arxiv.org/abs/2606.05342 | external | arxiv.org | Microsoft Research benchmark of 100 tasks across 10 synthetic web environments that evaluates long-running monitoring agents on whether they wait or act appropriately, scoring task completion, response speed, and resource efficiency. | Microsoft Research benchmark of 100 tasks across 10 synthetic web environments that evaluates long-running monitoring agents on whether they wait or act appropriately, scoring task completion, response speed, and resource efficiency. | Microsoft Research benchmark of 100 tasks across 10 synthetic web environments that evaluates long-running monitoring agents on whether they wait or act appropriately, scoring task completion, response speed, and resource efficiency. | The work turns loop quality into a measurable task or score. Microsoft Research benchmark of 100 tasks across 10 synthetic web environments that evaluates long-running monitoring agents on whether they wait or act appropriately, scoring task completion, response speed, and resource efficiency. | Use SentinelBench: A Benchmark for Long-Running Monitoring Agents to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,493 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1493 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;exit | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2606.05342 | [2606.05342] SentinelBench: A Benchmark for Long-Running Monitoring Agents | AI agents are increasingly asked to carry out work that spans minutes, hours, or longer. Yet the default model of agent behavior is continuous action: issuing tool calls, refreshing pages, searching for alternatives, or otherwise trying to force progress. This is the wrong approach for many long-running tasks, which ar... | Matheus Kunzler Maldaner; Adam Fourney; Amanda Swearngin; Hussein Mozannar; Gagan Bansal; Maya Murad; Rafah Hosn; Saleema Amershi | 2026-06-03 | 2026 | arXiv | arXiv | 18 pages, 16 figures | cs.AI | arxiv-api | 2606.05342 | 2026-09-08T16:27:48 | ||||||||
ale-0787 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SWE-Together: Evaluating Coding Agents in Interactive User Sessions | https://arxiv.org/abs/2606.29957 | external | arxiv.org | Multi-session coding benchmark of 109 repository-level tasks reconstructed from 11,260 recorded user-agent sessions, replayed with an LLM user simulator and scored on final correctness and the number of corrective feedback turns. | Multi-session coding benchmark of 109 repository-level tasks reconstructed from 11,260 recorded user-agent sessions, replayed with an LLM user simulator and scored on final correctness and the number of corrective feedback turns. | Multi-session coding benchmark of 109 repository-level tasks reconstructed from 11,260 recorded user-agent sessions, replayed with an LLM user simulator and scored on final correctness and the number of corrective feedback turns. | The work turns loop quality into a measurable task or score. Multi-session coding benchmark of 109 repository-level tasks reconstructed from 11,260 recorded user-agent sessions, replayed with an LLM user simulator and scored on final correctness and the number of corrective feedback turns. | Use SWE-Together: Evaluating Coding Agents in Interactive User Sessions to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,494 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1494 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2606.29957 | [2606.29957] SWE-Together: Evaluating Coding Agents in Interactive User Sessions | Most coding-agent benchmarks are static: an agent receives a complete task description up front and is judged only by its final code. Real coding assistance is interactive, with users clarifying goals, adding constraints, and correcting mistakes over multiple turns. We introduce SWE-Together, a multi-turn benchmark rec... | Yifan Wu; Zhuokai Zhao; Songlin Li; Ho Hin Lee; Jiacheng Zhu; Shirley Wu; Tianhe Yu; Serena Li; Lizhu Zhang; Xiangjun Fan; Shengzhi Li | 2026-06-29 | 2026 | arXiv | arXiv | cs.SE | arxiv-api | 2606.29957 | 2026-09-08T16:27:48 | |||||||||
ale-0788 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | The Long-Horizon Task Mirage? Diagnosing Where and Why Agentic Systems Break | https://arxiv.org/abs/2604.11978 | external | arxiv.org | Cross-domain diagnostic benchmark that scales task horizon through depth and breadth extension, then attributes failures across 3,100+ agent trajectories to a seven-category taxonomy via a trajectory-grounded LLM judge validated against human annotation. | Cross-domain diagnostic benchmark that scales task horizon through depth and breadth extension, then attributes failures across 3,100+ agent trajectories to a seven-category taxonomy via a trajectory-grounded LLM judge validated against human annotation. | Cross-domain diagnostic benchmark that scales task horizon through depth and breadth extension, then attributes failures across 3,100+ agent trajectories to a seven-category taxonomy via a trajectory-grounded LLM judge validated against human annotation. | The work turns loop quality into a measurable task or score. Cross-domain diagnostic benchmark that scales task horizon through depth and breadth extension, then attributes failures across 3,100+ agent trajectories to a seven-category taxonomy via a trajectory-grounded LLM judge validated against human annotation. | Use The Long-Horizon Task Mirage? Diagnosing Where and Why Agentic Systems Break to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,495 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1495 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;escalation | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2604.11978 | [2604.11978] The Long-Horizon Task Mirage? Diagnosing Where and Why Agentic Systems Break | Large language model (LLM) agents perform strongly on short- and mid-horizon tasks, but often break down on long-horizon tasks that require extended, interdependent action sequences. Despite rapid progress in agentic systems, these long-horizon failures remain poorly characterized, hindering principled diagnosis and co... | Xinyu Jessica Wang; Haoyue Bai; Yiyou Sun; Haorui Wang; Shuibai Zhang; Wenjie Hu; Mya Schroder; Bilge Mutlu; Dawn Song; Robert D Nowak | 2026-04-13 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2604.11978 | 2026-09-08T16:27:48 | |||||||||
ale-0789 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | Beyond pass@1: A Reliability Science Framework for Long-Horizon LLM Agents | https://arxiv.org/abs/2603.29231 | external | arxiv.org | Reliability metrics for long-horizon agents (reliability decay, variance amplification, graceful degradation, meltdown onset) measured over 23,392 episodes across 10 models, showing capability and reliability rankings diverge as tasks lengthen. | Reliability metrics for long-horizon agents (reliability decay, variance amplification, graceful degradation, meltdown onset) measured over 23,392 episodes across 10 models, showing capability and reliability rankings diverge as tasks lengthen. | Reliability metrics for long-horizon agents (reliability decay, variance amplification, graceful degradation, meltdown onset) measured over 23,392 episodes across 10 models, showing capability and reliability rankings diverge as tasks lengthen. | The work targets tasks that exceed a single context window or prompt session. Reliability metrics for long-horizon agents (reliability decay, variance amplification, graceful degradation, meltdown onset) measured over 23,392 episodes across 10 models, showing capability and reliability rankings diverge as tasks lengthe... | Use Beyond pass@1: A Reliability Science Framework for Long-Horizon LLM Agents to measure progress and gate completion with repeatable evidence. | Research source arXiv:2603.29231; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,496 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1496 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2603.29231 | [2603.29231] Beyond pass@1: A Reliability Science Framework for Long-Horizon LLM Agents | Existing benchmarks measure capability -- whether a model succeeds on a single attempt -- but production deployments require reliability -- consistent success across repeated attempts on tasks of varying duration. We show these properties diverge systematically as task duration grows, and that pass@1 on short tasks is ... | Aaditya Khanal; Yangyang Tao; Junxiu Zhou | 2026-03-31 | 2026 | arXiv | arXiv | 23 pages, 4 figures | cs.AI | arxiv-api | 2603.29231 | 2026-09-08T16:27:48 | ||||||||
ale-0790 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | SEAGym: An Evaluation Environment for Self-Evolving LLM Agents | https://arxiv.org/abs/2606.17546 | external | arxiv.org | Evaluation environment that measures whether a self-evolving agent's modifications to prompts, memory, and tools generalize to held-out tasks, using train, validation, and test splits and cost metrics on Terminal-Bench 2.0 and HLE. | Evaluation environment that measures whether a self-evolving agent's modifications to prompts, memory, and tools generalize to held-out tasks, using train, validation, and test splits and cost metrics on Terminal-Bench 2.0 and HLE. | Evaluation environment that measures whether a self-evolving agent's modifications to prompts, memory, and tools generalize to held-out tasks, using train, validation, and test splits and cost metrics on Terminal-Bench 2.0 and HLE. | Evaluation data is used as the feedback signal for improving loop behavior. Evaluation environment that measures whether a self-evolving agent's modifications to prompts, memory, and tools generalize to held-out tasks, using train, validation, and test splits and cost metrics on Terminal-Bench 2.0 and HLE. | Use SEAGym: An Evaluation Environment for Self-Evolving LLM Agents to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,497 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1497 | Verify | verify | Gate progress with tests, evals, and evidence. | workspace;context;verification;budget | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2606.17546 | [2606.17546] SEAGym: An Evaluation Environment for Self-Evolving LLM Agents | Self-evolving LLM-based agents improve mainly by changing their agent harness: the structured execution layer around a base model, including prompts, memory, tools, middleware, runtime state, and the model-tool interaction loop. Existing evaluations often reduce this process to isolated task scores or a single sequenti... | Congjie Zheng; Chuanyi Xue; Bin Liang; Jun Yang; Changshui Zhang | 2026-06-16 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2606.17546 | 2026-09-08T16:27:48 | |||||||||
ale-0791 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions | https://arxiv.org/abs/2605.24110 | external | arxiv.org | Benchmark of 26 evolving coding tasks across 227 evaluation rounds using cumulative executable tests to check that agents keep prior requirements working as specifications change, with top agents reaching only about 50% on multi-turn success metrics. | Benchmark of 26 evolving coding tasks across 227 evaluation rounds using cumulative executable tests to check that agents keep prior requirements working as specifications change, with top agents reaching only about 50% on multi-turn success metrics. | Benchmark of 26 evolving coding tasks across 227 evaluation rounds using cumulative executable tests to check that agents keep prior requirements working as specifications change, with top agents reaching only about 50% on multi-turn success metrics. | Evaluation data is used as the feedback signal for improving loop behavior. Benchmark of 26 evolving coding tasks across 227 evaluation rounds using cumulative executable tests to check that agents keep prior requirements working as specifications change, with top agents reaching only about 50% on multi-turn success me... | Use EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,498 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1498 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2605.24110 | [2605.24110] EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions | Coding agents are increasingly used as iterative development partners, but most benchmarks still evaluate one specification followed by one final assessment. This leaves out a basic question: can an agent keep its own codebase working as requirements change? We introduce EvoCode-Bench, a benchmark of 26 stateful coding... | Haiyang Shen; Xuanzhong Chen; Wendong Xu; Yun Ma; Liang Chen; Kuan Li | 2026-05-22 | 2026 | arXiv | arXiv | Work in Progress; 32 pages, 10 figures, preprint | cs.AI | arxiv-api | 2605.24110 | 2026-09-08T16:27:48 | ||||||||
ale-0792 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | On the Reliability of Computer Use Agents | https://arxiv.org/abs/2604.17849 | external | arxiv.org | Repeated-execution study on OSWorld decomposing why computer-use agents fail tasks they previously completed, separating execution stochasticity, task-specification ambiguity, and behavioral variability as distinct causes of unreliability. | Repeated-execution study on OSWorld decomposing why computer-use agents fail tasks they previously completed, separating execution stochasticity, task-specification ambiguity, and behavioral variability as distinct causes of unreliability. | Repeated-execution study on OSWorld decomposing why computer-use agents fail tasks they previously completed, separating execution stochasticity, task-specification ambiguity, and behavioral variability as distinct causes of unreliability. | Links loop design to measurable tasks where progress and failure can be compared. Repeated-execution study on OSWorld decomposing why computer-use agents fail tasks they previously completed, separating execution stochasticity, task-specification ambiguity, and behavioral variability as distinct causes of unreliability... | Use On the Reliability of Computer Use Agents to measure progress and gate completion with repeatable evidence. | Research source arXiv:2604.17849; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,499 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1499 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2604.17849 | [2604.17849] On the Reliability of Computer Use Agents | Computer-use agents have rapidly improved on real-world tasks such as web navigation, desktop automation, and software interaction, in some cases surpassing human performance. Yet even when the task and model are unchanged, an agent that succeeds once may fail on a repeated execution of the same task. This raises a fun... | Gonzalo Gonzalez-Pumariega; Saaket Agashe; Jiachen Yang; Ang Li; Xin Eric Wang | 2026-04-20 | 2026 | arXiv | arXiv | 33 pages, 3 figures, 4 tables | cs.AI | arxiv-api | 2604.17849 | 2026-09-08T16:27:48 | ||||||||
ale-0793 | Benchmarks And Evaluation | benchmarks-and-evaluation | Paper | 📄 | AgentLens: Revealing the Lucky Pass Problem in SWE-Agent Evaluation | https://arxiv.org/abs/2605.12925 | external | arxiv.org | Grades 2,614 SWE-agent trajectories across eight models to show that 10.7% of passing trajectories in its 1,815-trajectory evaluation subset are lucky trial-and-error successes, replacing binary pass/fail with process-quality tiers that shift model rankings. | Grades 2,614 SWE-agent trajectories across eight models to show that 10.7% of passing trajectories in its 1,815-trajectory evaluation subset are lucky trial-and-error successes, replacing binary pass/fail with process-quality tiers that shift model rankings. | Grades 2,614 SWE-agent trajectories across eight models to show that 10.7% of passing trajectories in its 1,815-trajectory evaluation subset are lucky trial-and-error successes, replacing binary pass/fail with process-quality tiers that shift model rankings. | Evaluation data is used as the feedback signal for improving loop behavior. Grades 2,614 SWE-agent trajectories across eight models to show that 10.7% of passing trajectories in its 1,815-trajectory evaluation subset are lucky trial-and-error successes, replacing binary pass/fail with process-quality tiers that shift m... | Use AgentLens: Revealing the Lucky Pass Problem in SWE-Agent Evaluation to measure progress and gate completion with repeatable evidence. | Research source arXiv:2605.12925; inspect its method and evaluation before treating results as production evidence. | medium | README.md | 1,500 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1500 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | research-preprint | A | ok | https://arxiv.org/abs/2605.12925 | [2605.12925] AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation | Evaluation of software engineering (SWE) agents is dominated by a binary signal: whether the final patch passes the tests. This outcome-only view treats a principled solution and a chaotic trial-and-error process as equivalent. We show that this equivalence is empirically false. We evaluate 2,614 OpenHands trajectories... | Priyam Sahoo; Gaurav Mittal; Xiaomin Li; Shengjie Ma; Benjamin Steenhoek; Pingping Lin; Yu Hu | 2026-05-13 | 2026 | arXiv | arXiv | cs.SE | arxiv-api | 2605.12925 | 2026-09-08T16:27:48 | |||||||||
ale-0794 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | ORLoopBench: Solver-in-the-Loop Benchmarks for Self-Correction | https://arxiv.org/abs/2601.21008 | external | arxiv.org | Formalizes infeasible-model debugging as a solver-in-the-loop process where each action triggers solver re-execution and infeasibility recomputation, giving deterministic verification for iterative repair in operations research. | Formalizes infeasible-model debugging as a solver-in-the-loop process where each action triggers solver re-execution and infeasibility recomputation, giving deterministic verification for iterative repair in operations research. | Formalizes infeasible-model debugging as a solver-in-the-loop process where each action triggers solver re-execution and infeasibility recomputation, giving deterministic verification for iterative repair in operations research. | Verification is promoted from a final check to a loop-control signal. Formalizes infeasible-model debugging as a solver-in-the-loop process where each action triggers solver re-execution and infeasibility recomputation, giving deterministic verification for iterative repair in operations research. | Use ORLoopBench: Solver-in-the-Loop Benchmarks for Self-Correction to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,501 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1501 | Verify | verify | Gate progress with tests, evals, and evidence. | trigger;verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://openreview.net/pdf/16a0193aa4e71ffe6c921ac0081a66b525eea017.pdf | [2601.21008] ORLoopBench: Solver-in-the-Loop Benchmarks for Self-Correction and Behavioral Rationality in Operations Research | Operations Research practitioners debug infeasible models through an iterative process: inspecting Irreducible Infeasible Subsystems ( IIS), identifying constraint conflicts, and repairing formulations until feasibility is restored. Existing LLM benchmarks mostly treat OR as one-shot translation from problem descriptio... | Ruicheng Ao; David Simchi-Levi; Xinshang Wang | 2026 | 2026 | Proceedings of the 43rd International Conference on Machine Learning (ICML), PMLR 306 | PMLR | Published in Proceedings of the 43rd International Conference on Machine Learning (ICML), PMLR 306; the linked arXiv record remains available for open access. | cs.LG | PMLR camera-ready record | 2601.21008 | 2026-09-08T16:27:48 | ||||||||
ale-0795 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis | https://arxiv.org/abs/2605.30434 | external | arxiv.org | Benchmark of 68 real-world data-analysis tasks built from Kaggle notebooks spanning 2,225 interactive turns, finding that long-horizon errors account for 52-69% of agent failures and that maintaining a correct analytical state is the core bottleneck. | Benchmark of 68 real-world data-analysis tasks built from Kaggle notebooks spanning 2,225 interactive turns, finding that long-horizon errors account for 52-69% of agent failures and that maintaining a correct analytical state is the core bottleneck. | Benchmark of 68 real-world data-analysis tasks built from Kaggle notebooks spanning 2,225 interactive turns, finding that long-horizon errors account for 52-69% of agent failures and that maintaining a correct analytical state is the core bottleneck. | The work turns loop quality into a measurable task or score. Benchmark of 68 real-world data-analysis tasks built from Kaggle notebooks spanning 2,225 interactive turns, finding that long-horizon errors account for 52-69% of agent failures and that maintaining a correct analytical state is the core bottleneck. | Use LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,502 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1502 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;state | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2605.30434 | [2605.30434] LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis | Real-world data analysis is inherently iterative, yet existing benchmarks mostly evaluate isolated or short interactive tasks, leaving agents' ability to track evolving analytical context over long horizons untested. We introduce LongDS, a benchmark for long-horizon, multi-turn data analysis where agents must maintain,... | Kewei Xu; Xiaoben Lu; Shuofei Qiao; Zihan Ding; Haoming Xu; Lei Liang; Ningyu Zhang | 2026-05-28 | 2026 | arXiv | arXiv | ACL 2026 | cs.LG | arxiv-api | 2605.30434 | 2026-09-08T16:27:48 | ||||||||
ale-0796 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks | https://arxiv.org/abs/2602.16313 | external | arxiv.org | Multi-session benchmark of interdependent agentic tasks where agents must distill earlier sessions into memory and use it to guide later actions, showing that near-saturated scores on long-context memory benchmarks fail to transfer. | Multi-session benchmark of interdependent agentic tasks where agents must distill earlier sessions into memory and use it to guide later actions, showing that near-saturated scores on long-context memory benchmarks fail to transfer. | Multi-session benchmark of interdependent agentic tasks where agents must distill earlier sessions into memory and use it to guide later actions, showing that near-saturated scores on long-context memory benchmarks fail to transfer. | The work turns loop quality into a measurable task or score. Multi-session benchmark of interdependent agentic tasks where agents must distill earlier sessions into memory and use it to guide later actions, showing that near-saturated scores on long-context memory benchmarks fail to transfer. | Use MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,503 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1503 | Verify | verify | Gate progress with tests, evals, and evidence. | context;verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2602.16313 | [2602.16313] MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks | Existing evaluations of agents with memory typically assess memorization and action in isolation. One class of benchmarks evaluates memorization by testing recall of past conversations or text but fails to capture how memory is used to guide future decisions. Another class focuses on agents acting in single-session tas... | Zexue He; Yu Wang; Churan Zhi; Yuanzhe Hu; Tzu-Ping Chen; Lang Yin; Ze Chen; Tong Arthur Wu; Siru Ouyang; Zihan Wang; Jiaxin Pei; Julian McAuley; Yejin Choi; Alex Pentland | 2026-02-18 | 2026 | arXiv | arXiv | cs.CL | arxiv-api | 2602.16313 | 2026-09-08T16:27:48 | |||||||||
ale-0797 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | Momento: Evaluating Persistent Memory and Reasoning with Multi-Session Agentic Conversations | https://arxiv.org/abs/2606.00832 | external | arxiv.org | Benchmark for persistent, tool-mediated task completion across multiple sessions, finding that agents fail by treating prior-session history as current context instead of stale state that needs re-validation. | Benchmark for persistent, tool-mediated task completion across multiple sessions, finding that agents fail by treating prior-session history as current context instead of stale state that needs re-validation. | Benchmark for persistent, tool-mediated task completion across multiple sessions, finding that agents fail by treating prior-session history as current context instead of stale state that needs re-validation. | The work turns loop quality into a measurable task or score. Benchmark for persistent, tool-mediated task completion across multiple sessions, finding that agents fail by treating prior-session history as current context instead of stale state that needs re-validation. | Use Momento: Evaluating Persistent Memory and Reasoning with Multi-Session Agentic Conversations to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,504 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1504 | Verify | verify | Gate progress with tests, evals, and evidence. | workspace;context;verification;state;exit | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2606.00832 | [2606.00832] Momento: Evaluating Persistent Memory and Reasoning with Multi-Session Agentic Conversations | Recent advances in agentic AI have enabled agents to complete complex tasks through tool use, reasoning, and multi-step planning. Yet existing benchmarks evaluate agents within a single session, ignoring past actions, stated preferences, and prior decisions that agents must integrate to fulfill personalized user goals.... | Adril Putra Merin; David Anugraha; Ayu Purwarianti; Genta Indra Winata | 2026-05-30 | 2026 | arXiv | arXiv | Preprint | cs.CL | arxiv-api | 2606.00832 | 2026-09-08T16:27:48 | ||||||||
ale-0798 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | π-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows | https://arxiv.org/abs/2605.14678 | external | arxiv.org | Benchmark of 100 multi-turn tasks across 5 user personas with hidden intents, inter-task dependencies, and cross-session continuity, measuring agent proactivity separately from task completion in long-horizon trajectories. | Benchmark of 100 multi-turn tasks across 5 user personas with hidden intents, inter-task dependencies, and cross-session continuity, measuring agent proactivity separately from task completion in long-horizon trajectories. | Benchmark of 100 multi-turn tasks across 5 user personas with hidden intents, inter-task dependencies, and cross-session continuity, measuring agent proactivity separately from task completion in long-horizon trajectories. | The work turns loop quality into a measurable task or score. Benchmark of 100 multi-turn tasks across 5 user personas with hidden intents, inter-task dependencies, and cross-session continuity, measuring agent proactivity separately from task completion in long-horizon trajectories. | Use π-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,505 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1505 | Verify | verify | Gate progress with tests, evals, and evidence. | verification;exit | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2605.14678 | [2605.14678] $π$-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows | The rise of personal assistant agents, e.g., OpenClaw, highlights the growing potential of large language models to support users across everyday life and work. A core challenge in these settings is proactive assistance, since users often begin with underspecified requests and leave important needs, constraints, or pre... | Haoran Zhang; Luxin Xu; Zhilin Wang; Runquan Gui; Shunkai Zhang; Haodi Lei; Zihao He; Bingsu He; Chicheng Qin; Tong Zhu; Xiaoye Qu; Yang Yang; Yu Cheng; Yafu Li | 2026-05-14 | 2026 | arXiv | arXiv | 44 pages | cs.AI | arxiv-api | 2605.14678 | 2026-09-08T16:27:48 | ||||||||
ale-0799 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | Can LLM Agents Be CFOs? Benchmarking Long-Horizon Resource Allocation | https://arxiv.org/abs/2603.23638 | external | arxiv.org | A 132-month CFO simulation where agents repeat a monthly cycle of liquidity management, financial closings, and financing decisions with compounding state, and only 15.4% of trials survive the full horizon. | A 132-month CFO simulation where agents repeat a monthly cycle of liquidity management, financial closings, and financing decisions with compounding state, and only 15.4% of trials survive the full horizon. | A 132-month CFO simulation where agents repeat a monthly cycle of liquidity management, financial closings, and financing decisions with compounding state, and only 15.4% of trials survive the full horizon. | The work targets tasks that exceed a single context window or prompt session. A 132-month CFO simulation where agents repeat a monthly cycle of liquidity management, financial closings, and financing decisions with compounding state, and only 15.4% of trials survive the full horizon. | Use Can LLM Agents Be CFOs? Benchmarking Long-Horizon Resource Allocation to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,506 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1506 | Verify | verify | Gate progress with tests, evals, and evidence. | state | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2603.23638 | [2603.23638] Can LLM Agents Be CFOs? Benchmarking Long-Horizon Resource Allocation in an Uncertain Enterprise Environment | Large language model (LLM) agents are increasingly tested on complex tasks, but their ability to allocate scarce resources over long horizons remains unclear. Unlike reactive tasks with immediate feedback, this setting requires agents to make binding commitments under partial observability, delayed consequences, hard r... | Yi Han; Yan Wang; Lingfei Qian; Haohang Li; Yupeng Cao; Yueru He; Xueqing Peng; Nanhan Shen; Yitao Xu; Yankai Chen; Dongji Feng; Jimin Huang; Xue Liu; Jian-Yun Nie; Sophia Ananiadou | 2026-03-24 | 2026 | arXiv | arXiv | cs.AI | arxiv-api | 2603.23638 | 2026-09-08T16:27:48 | |||||||||
ale-0800 | Benchmarks And Evaluation | benchmarks-and-evaluation | Benchmark | 🧪 | EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer | https://arxiv.org/abs/2607.05202 | external | arxiv.org | Benchmark isolating whether agents transfer reusable procedures such as searching, debugging, and verification across episodes in four long-horizon domains linked by ability graphs, finding curated experience transfers but no automatic method yields consistent gains. | Benchmark isolating whether agents transfer reusable procedures such as searching, debugging, and verification across episodes in four long-horizon domains linked by ability graphs, finding curated experience transfers but no automatic method yields consistent gains. | Benchmark isolating whether agents transfer reusable procedures such as searching, debugging, and verification across episodes in four long-horizon domains linked by ability graphs, finding curated experience transfers but no automatic method yields consistent gains. | Control flow is represented as an inspectable graph rather than an opaque prompt loop. Benchmark isolating whether agents transfer reusable procedures such as searching, debugging, and verification across episodes in four long-horizon domains linked by ability graphs, finding curated experience transfers but no automat... | Use EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer to measure progress and gate completion with repeatable evidence. | Benchmark or leaderboard source with repeatable tasks or scores; compare systems only after checking setup and scope. | high | README.md | 1,507 | https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/README.md#L1507 | Verify | verify | Gate progress with tests, evals, and evidence. | verification | researcher;evaluator | evaluation | enabling | benchmark | A | ok | https://arxiv.org/abs/2607.05202 | [2607.05202] EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer | Agent self-evolution in long-horizon LLM systems is largely procedural: useful experience is not merely stored information, but reusable procedures for searching, debugging, and verification. Yet current evaluations do not isolate this form of transfer. Agent benchmarks test single-episode task solving; memory benchmar... | Xingze Gao; Chuanrui Hu; Hongda Chen; Pengfei Yao; Zhao Wang; Yi Bai; Zhengwei Wu; Yunyun Han; Xiaofeng Cong; Jie Gui; Yafeng Deng; Teng Li | 2026-07-06 | 2026 | arXiv | arXiv | 15 pages, 2 figures, 8 tables | cs.AI | arxiv-api | 2607.05202 | 2026-09-08T16:27:48 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.