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/meapri/orchestrate-codex/orchestratenpx skills add Meapri/orchestrate-codex --skill orchestrategit clone --depth 1 https://github.com/Meapri/orchestrate-codexWhat 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 | $0.00067 | $0.00730 |
| Opus 5 | $0.00034 | $0.00365 |
| Sonnet 5 | $0.00013 | $0.00146 |
| Haiku 4.5 | $0.00007 | $0.00073 |
Grade A, and why
orchestrate 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 2d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrate (Claude Code host)
You (Claude) are the orchestrator. The orchestrate-codex MCP supplies the routing
brief, prepares each delegated leaf call (latest model + context + policy), and guards
quality — but the plan is YOUR judgment: which sub-tasks exist, which model handles
each, and what you simply do yourself.
Host-driven flow
orchestrate_advise— routing brief: available leaves, latest confirmed model ids (leaflist_modelscatalogs are stale — trust the brief, not leaf defaults), strength guide, doc-class policies.- You plan the sub-tasks and assignments.
- Per delegation:
orchestrate_step { capability, instruction, doc_class, gather?, context?, write_task?, model? }→{tool, arguments, fallback_tools, verify_after}. - Invoke that leaf tool with the prepared arguments. In Claude Code the leaf MCP
servers may not be attached to the session — execute the call via the broker's
leaf client instead:
.venv/bin/python -c "from orchestrate_codex.leaf_client import ..."ororchestrate_run(autonomous, needs~/.orchestrate_codex/leaves.json). - On failure try the next
fallback_toolsentry. Ifverify_after, runorchestrate_verify { text, doc_class, project_root }and re-delegate on warnings. - Synthesize results yourself; pass earlier findings forward via
context.
Routing guide (defaults — override with judgment)
- Claude leaf (opus) — depth, architecture analysis, rigorous review. (You may also simply do this reasoning yourself — you ARE a Claude model; delegate when you want an independent pass or to preserve your own context.)
- Grok — fast, broad, genuinely independent second opinion.
- Gemini write leaf — final structured authoring (readme/pr/release/translate);
self-grounds durable facts. Takes
task/instruction/source_text, never rawprompt. - grounded_search — current/external facts with sources.
- image / review_diff / compare — image gen, git-diff review, multi-model comparison.
- Do yourself: planning, file edits, running commands/tests, simple transforms.
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.
- 2d ago First seen · 51 lines · 67 tokens per session scan A 71436157d5c3
orchestrate is a skill published in the GitHub repository Meapri/orchestrate-codex (0 stars, last pushed 1mo ago), licensed MIT. It adds 67 tokens to every session and 730 once invoked, about $0.0003 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.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
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…