Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/ivy00johns/skill-madness/orchestrator-task-loopnpx skills add ivy00johns/Skill-Madness --skill orchestrator-task-loopgit clone --depth 1 https://github.com/ivy00johns/Skill-MadnessWrote 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.
[](https://agentmods.dev/skills/ivy00johns/skill-madness/orchestrator-task-loop)<a href="https://agentmods.dev/skills/ivy00johns/skill-madness/orchestrator-task-loop"><img src="https://agentmods.dev/badge/skills/ivy00johns/skill-madness/orchestrator-task-loop.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00193 | $0.03862 |
| Opus 5 | $0.00097 | $0.01931 |
| Sonnet 5 | $0.00039 | $0.00772 |
| Haiku 4.5 | $0.00019 | $0.00386 |
Grade A, and why
orchestrator-task-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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 240 lines — stays where its author put it; the contents beside it link to each section on GitHub.
orchestrator-task-loop
A configuration of [
loop-controller]. That skill supplies the loop machinery — primitive selection, the full guardrail stack, state externalization. This skill supplies the two things specific to draining the task list: a mechanical proof (every taskcompletedand passing itsTaskCompletedgate) and the assignment/supervision discipline that keeps the loop from declaring victory on a half-drained list. Readloop-controllerfor the guardrails; they're inherited, not repeated here.Why
disable-model-invocation: this loop spawns teammates and drives an entire task list to completion on its own — assigning, supervising, gating, and re-triggering its own stop. You want to type/orchestrator-task-loop(or have the orchestrator dispatch it) — not have Claude silently start an unbounded team-supervision loop because a task board happened to exist.Experimental. Agent Teams is disabled by default and gated behind
CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1, with known limitations (no session resumption for in-process teammates, lagging task-state propagation, slow shutdown). This is a prototype-behind-the-flag skill. If the flag is unset, it refuses to start and points you at the orchestrator's subagent/sequential path (see Step 1 + the degradation note).
The 5-part contract
| Part | This loop |
|---|---|
| trigger | A populated Agent Teams shared task list (tasks pending/in_progress, team spawned) and CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 — or an explicit /orchestrator-task-loop. Dispatched by the orchestrator after contracts exist and the board is seeded (one task per contract-bounded unit, blockedBy edges from the dependency graph). |
| action | One supervision pass: read the board → for every idle teammate (surfaced by TeammateIdle), assign the highest-priority pending task whose blockedBy set is fully completed → let teammates work in their own context windows and claim tasks → on completed, run that task's TaskCompleted gate (pass → done; fail → rolled back to pending with the failure as feedback) → surface blocked/ambiguous tasks for HITL. Assignment, not implementation — the loop routes and supervises; teammates do the work. |
| proof | The board is drained: every task completed AND each passed its TaskCompleted gate AND no blockedBy edge points at a non-completed task — observed in one full read. Default-FAIL: assume undrained until that whole-board predicate holds. (See "The proof".) |
| memory | The shared task list file is the durable state (task states + blockedBy + gate verdicts survive across passes and fresh contexts) — this loop's "feature-list JSON," default-FAIL. Plus a checkpoint commit per task, the team's peer-message log, and an optional PROGRESS.md roll-up. No cross-iteration state lives in the lead's conversation. |
| stop | board drained (success) OR hard iteration (pass) cap OR the 3-failure circuit breaker (board/task unchanged across 3 consecutive passes, or a task ping-pongs completed→gate-fail→pending ≥3×) OR enforced token/cost budget cap OR an HITL-blocked task with no unblocked work left. |
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 240 lines · 193 tokens per session scan A 43a22c2e3b32
orchestrator-task-loop is a skill published in the GitHub repository ivy00johns/Skill-Madness (11 stars, last pushed 26d ago), licensed MIT. It adds 193 tokens to every session and 3,862 once invoked, about $0.0010 per session on Opus 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
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.
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…
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…