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 skills add jhoblitt/rook-claude --skill rook-pr-takeovergit clone --depth 1 https://github.com/jhoblitt/rook-claudeWrote 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/jhoblitt/rook-claude/rook-pr-takeover)<a href="https://agentmods.dev/skills/jhoblitt/rook-claude/rook-pr-takeover"><img src="https://agentmods.dev/badge/skills/jhoblitt/rook-claude/rook-pr-takeover/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/jhoblitt/rook-claude/rook-pr-takeover"><img src="https://agentmods.dev/badge/skills/jhoblitt/rook-claude/rook-pr-takeover.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 | $0.00000 | $0.01716 |
| Opus 5 | $0.00000 | $0.00858 |
| Sonnet 5 | $0.00000 | $0.00343 |
| Haiku 4.5 | $0.00000 | $0.00172 |
Grade A, and why
rook-pr-takeover 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 7d 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR takeover — adopting an abandoned or unowned PR
A maintainer-responsibility transfer, invoked ONLY on the user's explicit
per-PR instruction ("take over #N", "adopt this", "supersede it"). The
reviewer output may flag takeover_candidate, but flagging is information;
the transfer is a human decision. Use cases: avoiding long interaction
cycles, an author who abandoned the PR or stopped responding to change
requests, or AI-burst PRs where no response is reasonably expected.
Mental model: the PR is treated as a bug report with a candidate patch attached. The maintainer becomes the owner of landing the fix; the original author keeps credit for finding and writing it.
Takeover only applies to PRs whose substance is worth landing. If the bug
was judged FABRICATED or the change is unwanted, the honest disposition is
close-with-explanation via normal triage (rook-triage) — not takeover.
Choosing the approach
- Adopt in place — when ONLY the PR title/description are deficient: code verified good, commit messages acceptable, no rebase or code changes needed. The existing PR stays open under the original author; you edit it directly.
- Supersede — open a replacement PR carrying the commits when commit messages need correction, review found code blockers to fix, the branch has conflicts, or the maintainer wants to carry the change with modifications.
- Commit messages cannot be fixed by adopting in place — that would mean rewriting the author's fork branch (possible only with "allow edits by maintainers", and out of bounds for this flow). Wrong commit messages → supersede.
Adopt in place — edit the existing PR (gh pr edit)
- Start from the review's verified narrative — the reviewer's
suggested_title/suggested_bodywhen present, else draft from the review rationale. - Title: accurate and conventional (type from
.commitlintrc.jsonmatching the change). - Body: the standard PR template — accurate mechanism narrative (what the
verification actually established, not the original claims), a truthful
checklist against the real diff, PendingReleaseNotes/backport
consideration — plus a transparency note at the end:
"Description rewritten by maintainer for accuracy; original report and
fix by @." and an AI-assistance disclosure for the rewrite per
ai-guidelines.md. - Show the user the exact final title and body; on approval,
gh pr edit <n> --title ... --body-file ... --add-assignee @me(sandbox disabled) — assigning the taken-over PR to the maintainer marks it as theirs to shepherd (best-effort; skip on permission failure). GitHub's edit history plus the note keep the intervention visible.
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.
- 7d ago Changed · +1 lines 3f0bb274d9cc
- 11d ago First seen · 119 lines · 0 tokens per session scan A 5988204ec6af
rook-pr-takeover is a skill published in the GitHub repository jhoblitt/rook-claude (4 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,716 tokens. 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
cad-reviewer-contract
Internal role contract, preloaded into every cad-reviewer rung agent. Not a user command.
pr-deslop
Use when cleaning AI slop, verbose commit messages, brittle references, or low-value changes from a branch before review.
slop-scan
Use when scanning tracked repository files for AI slop, verbosity, brittle references, or low-value contributions.
seal
Ships a feature — simplify pass, certify, commit, PR, CI monitoring, review loop, merge. Use when a feature branch is verified and ready to ship.
weave-fix-review
Fix weave review findings — validate, add test coverage, fix, and commit each as atomic changes.
action
Use when screened review findings should be fixed on the current branch, usually as one verified commit per finding.