StackMap
Subscribe

cwc-long-running-agents vs loop-engineering

Anthropic's harness primitives for long-running Claude agents: default-FAIL evidence gates, a fresh-context evaluator subagent and handoff hooks — each one standalone, readable file. — versus — Reference repo plus npm CLIs (loop-init/audit/cost) for loop engineering: designing scheduled, gated control loops that prompt and orchestrate AI coding agents — Grok, Claude Code, Codex — over time.

The curated verdict

Both are reference repos for keeping agents productive over long horizons: loop-engineering designs the control loops, Anthropic's primitives enforce evidence and evaluation inside Claude Code hooks.

cwc-long-running-agentsloop-engineering
Stars5798.0k
Forks601.1k
LanguageShellJavaScript
LicenseApache-2.0MIT
Last activity2 months ago3 days ago
Topicscodingcoding, orchestration
Curated connections38

cwc-long-running-agents — the curator's take

Read it, don't run it: an official worked example of WHY long runs fail (agents grading their own work, claiming success without evidence, losing state between sessions) with one hook per failure mode. The default-FAIL contract is the idea worth stealing. Explicitly an event demo — unmaintained, not accepting contributions — so copy the patterns into your own harness rather than depending on the repo.

loop-engineering — the curator's take

The methodology layer for autonomous coding agents — read it once you've stopped hand-prompting and want to design the loop that prompts the agent instead: scheduled triage, worktree isolation, maker/checker sub-agents, MCP connectors, human gates, and a phased L1-report → L3-unattended rollout. Ships 7 clone-and-run patterns, tool-agnostic starters (Grok/Claude Code/Codex/OpenCode), and CLIs that score loop-readiness (loop-audit), estimate token spend (loop-cost) and scaffold state/budget (loop-init). It's a patterns + tooling reference, NOT a runtime — it won't execute or host your agents, so bring your own harness (an alook-style AI-company layer or a squid-style pipeline). Overkill if you just want one-shot interactive help; heed its own caveat that unattended loops make unattended mistakes, so verification stays on you.