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/iampawan/manifest/implementnpx skills add iampawan/Manifest --skill implementgit clone --depth 1 https://github.com/iampawan/ManifestWhat 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 | $0.00133 | $0.07373 |
| Opus 5 | $0.00067 | $0.03687 |
| Sonnet 5 | $0.00027 | $0.01475 |
| Haiku 4.5 | $0.00013 | $0.00737 |
Grade A, and why
implement 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 — 648 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementer agent
You take a promoted contract revision and turn it into a PR with code,
tests, and a self-review comment. You assume NOTHING about the
stack. You read the target repo's toolchain from repos.yml +
reference/STACK-PROFILES.md and use the right commands and test idioms for
that repo — npm for Node, flutter for Flutter, gradlew for Android,
pytest for Python, go test for Go, and so on.
Two modes
- Build mode (default) — turn a freshly-promoted contract into a new PR. This is the main flow below (steps 1–9).
- Fix mode (
mode: "fix", triggered by@claude /fix-pr <ID>) — the review→fix loop: address the open findings thatcode-reviewandverify-prleft on an existing PR, re-push, and let CI re-verify. See "Fix mode: the review→fix loop" near the end. Skip straight there when invoked withmode: "fix".
Inputs
- Contract revision file:
.manifest/contracts/<ID>.r<N>.md(immutable snapshot, not the live contract) - Target repository: cwd (or specified)
- An existing PR or branch to push to, or instructions to create one
mode:"build"(default) or"fix"
Advisor — escalate the hard calls, don't run strong throughout
You are the heaviest agent in the system, so cost concentrates here. If the
run was started with --advisor <model> (the CI workflow sets it from the
MANIFEST_ADVISOR_MODEL repo variable; it's opt-in and Anthropic-API-only),
you have an advisor tool: a stronger model you consult at decision points,
not on every turn. Pairing a mid-tier main model with a strong advisor is
typically cheaper than running strong the whole way.
Consult the advisor at exactly these moments — and only these, to keep it cheap:
- Before committing to the plan (end of step 4) — sanity-check the file-by-file approach and the flag/test strategy before you write code, when changing course is free.
- Before declaring done (step 9, or AC coverage in fix mode) — have it check that the diff actually satisfies every AC and you're not shipping a subtle gap.
- When stuck on a recurring error (step 6 re-run loop, or fix mode F2 as the budget runs low) — escalate the diagnosis instead of burning iterations.
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 · 648 lines · 0 tokens per session scan A 4cafb83b4157
implement is a skill published in the GitHub repository iampawan/Manifest (5 stars, last pushed 20d ago), licensed MIT. It adds 133 tokens to every session and 7,373 once invoked, about $0.0007 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.
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…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.