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 richkuo/rk-skills --skill fix-pr-review-loopgit clone --depth 1 https://github.com/richkuo/rk-skillsWrote 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/richkuo/rk-skills/fix-pr-review-loop)<a href="https://agentmods.dev/skills/richkuo/rk-skills/fix-pr-review-loop"><img src="https://agentmods.dev/badge/skills/richkuo/rk-skills/fix-pr-review-loop/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/richkuo/rk-skills/fix-pr-review-loop"><img src="https://agentmods.dev/badge/skills/richkuo/rk-skills/fix-pr-review-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00132 | $0.03896 |
| Opus 5.5 | $0.00053 | $0.01558 |
| Sonnet 5.5 | $0.00026 | $0.00779 |
| Haiku 4.5 | $0.00013 | $0.00390 |
Grade A, and why
fix-pr-review-loop 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 20d 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
fix-pr-review-loop
Resolve the latest review with fix-pr-review, wait for the bot's re-review, repeat. This skill owns the step 3 stop rules and the step 5 table. work-on-issue-loop runs steps 2 to 4 and reuses step 5's table with deltas; edit its step 4 with that table.
Input
Nothing (the current branch's PR via gh pr view), or #<N> / <N> / URL / owner/repo#N.
Steps
1. Resolve the PR and establish the starting state
Run gh pr view <N|--> --json number,headRefName,headRepositoryOwner,baseRefName,url,state,isDraft; merged or closed: stop and report. Compute review_count (0 on a fresh PR) and pr_cycle_count per Round counts in skills/fix-pr-review/rereview-routing.md, which owns both counts and the pending trigger; a verification limitation there stops the loop: report the missing evidence, post nothing. Fetch feedback with fix-pr-review step 1's three-channel query (reviews, issue comments, inline threads). Take the first case that applies:
- Pending trigger present (per Round counts, naming the selected bot): recover it; never post another. One naming the other bot: stop and report.
- Unaddressed feedback present (a completed bot round newer than any disposition comment, or trusted feedback failing fix-pr-review step 1's bare-LGTM check): step 3, no trigger.
- Stale or failed trigger (per Round counts, naming the selected bot: stale, or its round has no verdict and its answering review run finished without one; a queued or in-progress re-run keeps it pending): repost it. A round with a verdict never takes this case; a note from any other run never makes a trigger failed. After a disposition comment, repost the trigger's exact line; else it was cycle 1: post the first-review trigger.
- Nothing to act on yet (fresh PR, only your own disposition or trigger comments, only untrusted feedback, or only a trusted bare
LGTMor empty approval): post the first-review trigger as its own one-line comment,gh pr comment <N> --body "<trigger>".
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.
- 20d ago Changed b19ef087aa3a
- 22d ago Changed f5c1dfb77c0f
- 24d ago Changed 3f3b53ac1676
- 27d ago Changed · +8 lines 66ed3c14f013
- 29d ago Changed · +24 lines c749ec1acae0
- 1mo ago Changed 106fd14023f8
- 1mo ago Changed · -8 lines acc980203ea6
- 1mo ago Changed 83db0178ae21
- 1mo ago Changed · -43 lines · -66 tokens per session 93e7418e205c
- 1mo ago First seen · 133 lines · 198 tokens per session scan A baf98db1eb47
fix-pr-review-loop is a skill published in the GitHub repository richkuo/rk-skills (50 stars, last pushed yesterday), licensed MIT. It adds 132 tokens to every session and 3,896 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
verification-gate
You MUST use this when work is about to be called complete, done, fixed, ready, or mergeable - before a completion claim, a merge or pull request, a branch finish, or a handoff report.
review-request
You MUST use this when a completed or partially committed implementation needs an independent code review against explicit requirements, a defined task scope, and the actual Git diff, including committed and working-tree changes.
address-github-comments
Use when you need to address review or issue comments on an open GitHub Pull Request using the gh CLI.
review-it
Code review closeout for Claude Code, Codex, OpenCode, DeepSeek TUI, and Antigravity CLI: local dirty changes, branch vs main, parallel tests.
check
Reviews diffs, PRs, release readiness, and publishing follow-through. Use when asked to review, triage issues or PRs, or ship. Not for debugging root causes or prose.
antfu-create-pr
Create a reviewable GitHub pull request from the current branch with a Conventional Commits title, a concise evidence-based body, and before/after screenshots for UI changes. Use when asked to open, create, publish, or prepare a PR.