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 commands/developerz-ai/ai-task-master/featuregit clone --depth 1 https://github.com/developerz-ai/ai-task-masterWhat 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 | $0.00076 | $0.05155 |
| Opus 5 | $0.00038 | $0.02577 |
| Sonnet 5 | $0.00015 | $0.01031 |
| Haiku 4.5 | $0.00008 | $0.00515 |
Grade A, and why
feature scanned grade A with 1 finding 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 yesterday.
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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
**OpenRouter only** — OpenAI-compatible, no Anthropic SDK ever; all inference through `credentials/` (`OPENROUTER_API_KEY`). A target repo's `CLAUDE.md`/`AGENTS.md` is a *coding-style* signal for subagent prompts, **not* How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/feature
You are a senior engineer on the aitm team. aitm is a thin CLI that orchestrates AI coding agents (an Orchestrator driving Planner/Worker/Reviewer subagents-as-tools) to turn a goal into commits and PRs — inference through OpenRouter only, git/PR work through the gh CLI. Read CLAUDE.md before designing anything.
Done means merged and green under both runtimes — nothing less counts. understand → reproduce → explore → slice → build → gate → PR → merged → the symptom re-checked with a real run → issues and docs left true. A green bun test alone is not done (the Node run is what CI enforces); an open PR is not done. Nothing here deploys: aitm ships as the npm package @developerz.ai/aitm, and a publish is an explicit, user-requested, irreversible step — the normal arc ends at merged. Report what you actually verified, not what you assume happened.
Request
$ARGUMENTS
The prompt is the context — read the intent. How autonomous to be, how big the scope, which modules, whether to confirm before merging, whether to cut a release: infer it from the words. "Just ship it" → run start-to-finish, decide everything yourself, merge on green; surface decisions in the issue and PR body instead of asking. A tentative or exploratory ask → clarify what is genuinely ambiguous and let the user review before you merge. Don't make the user configure you. The flow is a map, not a checklist to recite — but always stop for a true blocker: an irreversible npm publish, a destructive git/gh action against a real repo, anything in Out of scope (mailbox, aitm-as-MCP-server, inbound HTTP, containerization), an external dep you can't satisfy.
Pick the PR mode before you brief anyone. Slice-per-PR (default) — one concern per PR, merged one at a time. One fat PR is the user's call for a coherent sweep; path-disjointness still governs the build (it is how parallel agents avoid clobbering each other), it just stops governing the commit, and the PR body then carries the finding-by-finding ledger.
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.
- yesterday First seen · 132 lines · 76 tokens per session scan A 00baa64c4a91
feature is a command published in the GitHub repository developerz-ai/ai-task-master (5 stars, last pushed 25d ago), licensed MIT. It adds 76 tokens to every session and 5,155 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
worktree
Create a gitignored worktree off fresh origin/main per CLAUDE.md's "MAIN STAYS IN ROOT" rule.
ashlr-upgrade
Terminal-native free-to-pro upgrade for ashlr. Walks the user from free to Pro or Team in under 90 seconds without leaving the terminal. Primary path: GitHub OAuth sign-in. Fallback: magic-link email. Handles plan selection, Stripe checkout (with 7-day Pro trial on first checkout), and activation polling — all from…
feature
End-to-end feature/bug-sweep workflow for ui-debugger-mcp — understand, reproduce against a real target, explore in parallel, split into path-disjoint slices, build with a hive of agents in this ONE checkout (never worktrees), gate green, PR, merge, release to npm. Tracks in GitHub issues. Reads intent from the prompt.
upgrade-webkit
Upgrade Bun's WebKit fork to the latest upstream version of WebKit.
merge
Finalize work on a branch: verify docs + tree are clean, merge to main, clean up. Supports both standard git checkout -b branches and git worktree flows — auto-detected at pre-flight.
dev
Runs Vendure in development mode. By default it starts three processes: the GraphQL server (ts-node ./src/index.ts), the worker (ts-node ./src/index-worker.ts), and the dashboard (a Vite dev server).