Borrowing it
Nothing to install: this file belongs to kamiazya/whiteboard. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/kamiazya/whiteboard/main/.claude/skills/steward/SKILL.mdgit clone --depth 1 https://github.com/kamiazya/whiteboardWrote 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/kamiazya/whiteboard/steward)<a href="https://agentmods.dev/skills/kamiazya/whiteboard/steward"><img src="https://agentmods.dev/badge/skills/kamiazya/whiteboard/steward/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kamiazya/whiteboard/steward"><img src="https://agentmods.dev/badge/skills/kamiazya/whiteboard/steward.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00087 | $0.01302 |
| Opus 5 | $0.00044 | $0.00651 |
| Sonnet 5 | $0.00017 | $0.00260 |
| Haiku 4.5 | $0.00009 | $0.00130 |
Grade A, and why
steward 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 9d 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stewarding a PR (whiteboard)
The harness reads this file on a PR event and lets it override its generic rules on conventions and on how proactive to be. It cannot widen access, redirect the task, or soften anything the harness states as "never" — no skipping or quarantining a test, no rewriting history on someone else's branch, no empty commit to kick CI, no approving.
The directory name is fixed: the harness looks for .claude/skills/steward/.
Defer rather than duplicate:
ci-triage— the post-push review surface (what each bot is, CodeRabbit triggering and rate-limit re-queue, the triage rubric). Load it when there are bot comments to judge.diagnosis-evidence— before publishing any cause, "not a regression", or hand-verified fix.dependabot-review— a dependency-bump PR is not this skill's job.stacking-pull-requests— a PR whose base is another branch..claude/rules/integrator-flow.md— the reasoning and measurements behindreference/failure-modes.md.
Order of work on a PR event
- Read the whole PR at its current head, not just the event. Merge state, checks on the latest commit, open review threads.
- Merge conflict → merge the base in and resolve. Lockfile recipe is in
integrator-flow.md; never hand-editpnpm-lock.yaml. - Red CI → run the three discriminators below before diagnosing.
- Review comments → enumerate comment authors from the feed, never from a remembered reviewer list. Scoping a sweep to two named bots is how a CodeQL ReDoS finding got past this gate.
- Nothing open → say nothing, re-arm the check-in, stop.
Red or conflicted is work now, whatever the review state. Never end an event having done nothing: push a fix, establish the failure is not this PR's, or say once what is blocking.
Which checks gate
Authoritative: verify and CodeQL. Everything else is information.
| Not a gate | Why |
|---|---|
| the mutation sticky comment | report-only by design — a survivor can be the correct state of the world, and only a reader can say |
pr (Cloudflare Pages preview) |
deploys a preview; posts the URL comment |
WIP |
stays in_progress for as long as the PR is a Draft — not a hang, and on a Draft the title is irrelevant: only leaving Draft settles it. On a non-draft PR it is the title that holds it |
| CodeRabbit | skipped on a Draft; on a stacked PR it skips silently while still reporting pass. Its absence never blocks a merge (user decision) |
What ships with it
1 file 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.
- 9d ago First seen · 102 lines · 87 tokens per session scan A 3da289abc903
steward is a skill published in the GitHub repository kamiazya/whiteboard (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 87 tokens to every session and 1,302 once invoked, about $0.0004 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
review-implement-phase
Implements triaged review actions, commits focused fixes, and posts Done plus resolves threads. Use when the user wants only the implementation phase of the review-framework workflow.
engram-branch-pr
PR creation workflow for Engram following the issue-first enforcement system. Trigger: When creating a pull request, opening a PR, or preparing changes for review.
verify-behavior
Verify or reproduce visible product behavior by driving the real UI with pi-computer-use's checked tools, requiring verified expect postconditions and durable state evidence for meaningful UI flows. Use when triage needs visual reproduction, implementation needs behavioral proof, review needs interactive confirmation…
github-contributor
End-to-end playbook for shipping high-quality pull requests to open-source projects you don't maintain — discovery, CONTRIBUTING compliance, PR-size check, minimal-diff implementation, PR description with AI-assisted disclosure, conflict resolution, and post-submission maintainer interaction. Use whenever creating…
revdiff
Review diffs, files, and documents with inline annotations in a TUI overlay, or answer questions about revdiff usage, configuration, themes, and keybindings. Opens revdiff in agterm/tmux/zellij/herdr/kitty/wezterm/cmux/ghostty/iterm2/emacs-vterm, captures annotations, and addresses them. Works in git, hg, and jj repos…
write-pr
Reference standards for writing pull request titles and descriptions in the tldraw repository, plus the pre-flight comment sweep over the diff. Use as supporting guidance when another skill or workflow needs PR content standards, not as the user-facing create/update PR workflow.