An automated, repeating improvement process for a specific agent-orchestrator codebase. It observes workers, pull requests, and workflows, measures results, records problems, and dispatches fixes over repeated cycles.

Use it to monitor an agent-orchestrator project, measure its zero-touch rate, diagnose gaps, create tracked work items, and send fixes to workers.

It provides a defined way to find friction in an agent system and continue working through issues without pausing after each cycle. It also sets conditions for stopping.

How jleechanorg/agent-orchestrator-ts configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to jleechanorg/agent-orchestrator-ts. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/jleechanorg/agent-orchestrator-ts/main/.claude/skills/eloop/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/jleechanorg/agent-orchestrator-ts

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/jleechanorg/agent-orchestrator-ts/eloop/github.svg)](https://agentmods.dev/skills/jleechanorg/agent-orchestrator-ts/eloop)

Your own site

<a href="https://agentmods.dev/skills/jleechanorg/agent-orchestrator-ts/eloop"><img src="https://agentmods.dev/badge/skills/jleechanorg/agent-orchestrator-ts/eloop/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/jleechanorg/agent-orchestrator-ts/eloop"><img src="https://agentmods.dev/badge/skills/jleechanorg/agent-orchestrator-ts/eloop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,555 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00055	$0.02555
Opus 5.5	$0.00022	$0.01022
Sonnet 5.5	$0.00011	$0.00511
Haiku 4.5	$0.00006	$0.00255

Measured 9d ago against content hash 9196e712ca85, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

evolve_loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/eloop/SKILL.md · 245 lines

How it starts

The opening of the file, as written. The whole thing — 245 lines — stays where its author put it; the contents beside it link to each section on GitHub.

...

Evolve Loop — Agent Orchestrator Fork

This is the repo-local override. For generalized harness-evolution principles that apply across all repos, also consult the user-scope eloop skill: ~/.claude/skills/evolve-loop/SKILL.md.

Purpose

Autonomous self-improving loop for the agent-orchestrator fork. Observes the AO ecosystem (workers, PRs, workflows), measures zero-touch rate, diagnoses friction, creates beads for gaps, dispatches fixes via /claw, and records everything. Runs via /loop 10m for max 12 hours.

Autonomous Continuation

After completing Phase 7, immediately start Phase 1 of the next cycle. Do not pause for confirmation between cycles.

The loop stops only when one of these is true:

User explicitly says stop or pause
12 hours elapsed since first cycle
Context window exceeds 90%
System is stable for 3 consecutive healthy cycles

Treat "keep going" or "until stable" as standing directives.

Adaptive Behavior

This loop is problem-driven.

Healthy cycle: Observe -> Measure -> Recap
Problem cycle: Observe -> Measure -> Diagnose -> Plan -> Record -> Fix -> Recap

Decision rules after Phase 2:

Zero-touch rate unchanged and above 20%, no new friction, all workers alive: skip to recap
Zero-touch rate below 20% for 3+ consecutive cycles: perform code-level diagnosis
New dead worker or new PR failure: run diagnose through fix
Worker stuck for 3 checks: kill and respawn, skip full diagnosis
Build broken on main: fix immediately

For chronic problems, read the automation code rather than just checking infrastructure.

Loop Body

Phase 1: Observe

Memory search first — Run /ms (memory_search) to pull prior context for tracked PRs and friction points:
```
/ms "open PRs jleechanorg/agent-orchestrator-ts"
/ms "open PRs jleechanclaw"
/ms "stuck PRs zero-touch"
/ms "bead bd- [recent]"
```
This surfaces what happened in previous cycles, what was already dispatched, and what blockers are known — so the loop doesn't repeat work or miss context.

Read the full file on GitHub · 245 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 245 lines · 55 tokens per session scan A 9196e712ca85

Subscribe to this mod's changes

evolve_loop is a skill published in the GitHub repository jleechanorg/agent-orchestrator-ts (8 stars, last pushed 3d ago), licensed MIT. It adds 55 tokens to every session and 2,555 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-02.

Analytics and metrics Building agents

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

agent-orchestrator-ts: Skill for Claude Code