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/darrencroton/ai-agent-coder/orchestratornpx skills add darrencroton/ai-agent-coder --skill orchestratorgit clone --depth 1 https://github.com/darrencroton/ai-agent-coderWrote 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/darrencroton/ai-agent-coder/orchestrator)<a href="https://agentmods.dev/skills/darrencroton/ai-agent-coder/orchestrator"><img src="https://agentmods.dev/badge/skills/darrencroton/ai-agent-coder/orchestrator.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 | $0.00098 | $0.02872 |
| Opus 5 | $0.00049 | $0.01436 |
| Sonnet 5 | $0.00020 | $0.00574 |
| Haiku 4.5 | $0.00010 | $0.00287 |
Grade A, and why
orchestrator 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 5d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrator
The orchestrator is a workflow, not a person or a model tier. Any supported harness can run it; any supported harness can be launched by it. The assistant directly handling the user's request is the Developer for that session. The Developer owns session management, planning, implementation, testing, verification, commits, acceptance decisions, and the final user-facing deliverable — that ownership never transfers to a delegate, no matter which harness is doing the delegating or which harness is launched.
A delegate is an external harness session the Developer launches to do bounded work outside the current session. Every delegate launches in one of two access modes, chosen by the request and constrained by policy:
read-only— an evidence source. It inspects code and artifacts, gathers evidence, verifies plans, performs drift audits, and performs code reviews. It never edits files, runs mutation-prone commands, performs Git or GitHub mutations, commits, or re-delegates.read-write— a bounded implementer. It may create, edit, and run commands to complete a task, but only inside the request's explicitauthorized_surfaceand never past itsnon_goals. Like a read-only delegate, it never performs Git or GitHub mutations, never commits, and never re-delegates: the Developer reviews its diff and commits it, exactly as it would for its own edits.
A read-only delegate is what the other skills in this repository call a Reviewer; the two names describe the same read-only, gate-free helper.
The Developer must critically evaluate every delegate result — a read-only delegate's report before trusting it as evidence, and a read-write delegate's diff before accepting it as if it were the Developer's own work. Running drift-audit against a read-write delegate's diff is the expected next step, the same as after any other implementation.
Use scripts/delegate_jobs.py for every delegate launch. Read references/delegate-contract.md before writing a request. The helper validates schema-v3 semantic requests, embeds required skills, composes the selected harness command, records normalized access evidence, manages process lifecycle, and extracts results.
Delegate artifacts default to .orchestrator/runs/. Override that root with ORCHESTRATOR_ARTIFACT_ROOT. Existing .ai-orchestrator/ evidence is historical and must not be read, migrated, or reinterpreted.
What ships with it
14 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.
- .gitignore 215 B
- README.md 3.3 KB
- references/claude-code.md 2.9 KB
- references/codex.md 4.7 KB
- references/copilot.md 2.7 KB
- references/delegate-contract.md 10 KB
- references/opencode.md 3.3 KB
- references/qwen.md 4.3 KB
- references/templates.md 4.9 KB
- scripts/delegate_contract.py 34 KB runs code
- scripts/delegate_jobs.py 55 KB runs code
- scripts/delegate_sessions.py 33 KB runs code
- tests/test_delegate_contract.py 82 KB runs code
- tests/test_delegate_sessions.py 26 KB runs code
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.
- 5d ago First seen · 130 lines · 98 tokens per session scan A f58f1464dce0
orchestrator is a skill published in the GitHub repository darrencroton/ai-agent-coder (2 stars, last pushed 4d ago), licensed MIT. It adds 98 tokens to every session and 2,872 once invoked, about $0.0005 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 2026-08-31.
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…