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/petekp/claude-code-setup/pr-descriptionnpx skills add petekp/claude-code-setup --skill pr-descriptiongit clone --depth 1 https://github.com/petekp/claude-code-setupWrote 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/petekp/claude-code-setup/pr-description)<a href="https://agentmods.dev/skills/petekp/claude-code-setup/pr-description"><img src="https://agentmods.dev/badge/skills/petekp/claude-code-setup/pr-description.svg" alt="Measured on agentmods" 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 | $0.00120 | $0.02330 |
| Opus 5 | $0.00060 | $0.01165 |
| Sonnet 5 | $0.00024 | $0.00466 |
| Haiku 4.5 | $0.00012 | $0.00233 |
Grade A, and why
pr-description 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Description
A reviewer opens a PR with four questions: what does this do, why now, where's the risk, and how do I know it works. Answer those four things and stop. Everything else — file-by-file narration, restated diffs, boilerplate sections — costs reviewer attention without answering anything.
The description's job is to make the review fast and the merge safe, not to prove effort. A great one is shorter than the reader expects and shows more than it tells.
Process: evidence first, prose last
Write nothing until you have looked at what the change actually does. The order matters — prose written before evidence turns into summary; prose written after evidence turns into captions.
- Read the real diff.
git diff <base>...HEAD(orgh pr diff). Note what a reviewer will find confusing or risky — that goes in the description. Skip what's mechanical — that doesn't. - Find out why. Check the branch name, linked issue, commit messages, and conversation context. If you genuinely can't tell why this change exists, ask — a wrong "why" is worse than none.
- Capture proof of behavior (next section). Run the thing. Do not describe behavior you haven't seen.
- Write the prose around the evidence. Then cut it by a third.
Evidence: match the proof to the change
Every claim about behavior gets one piece of evidence. Pick by change type:
| Change | Proof |
|---|---|
| UI change | Before/after screenshot — use the pr-screenshot-comparison skill, which handles capture, stitching, and gh --attach |
| CLI or script behavior | A real terminal transcript, trimmed to the relevant lines, in a fenced block |
| Output/data/API shape | Before and after of the actual payload or output, side by side or stacked, fenced |
| Config or option change | A two-column table: old value → new value, one row per option that changed |
| Structural change (moved/split files, new module) | A small ascii tree or diagram of just the affected part |
| Bug fix | The failing behavior (error message, wrong output) and the fixed behavior |
| Pure refactor | Proof of no behavior change: test run output, or identical before/after output |
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 · 118 lines · 120 tokens per session scan A 877efaf17d6c
pr-description is a skill published in the GitHub repository petekp/claude-code-setup (47 stars, last pushed 2d ago), licensed MIT. It adds 120 tokens to every session and 2,330 once invoked, about $0.0006 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 2026-09-04.
Other skills, from other repositories
github
GitHub via gh CLI: PRs, issues, reviews, repos, auth.
bootstrap-repo-analysis
First-time analysis of a repository with no prior reviewer outcomes. Crawl historical merged-PR review feedback with the gh CLI (plus any preloaded samples), extract the team's review norms, and synthesize the initial per-repo review-style prompt. Use this for a cold-start repo; use continual-learning instead once the…
baby-sit
Monitor a GitHub pull request until CI is green, diagnose failures, and rerun only evidence-backed flaky GitHub Actions jobs.
continual-learning
Nightly refinement of an existing per-repo review-style prompt using this reviewer's own finding outcomes. Read confirmed (resolved-by-commit / thumbs-up) and dismissed (thumbs-down) findings, promote the bug patterns the team actually fixes, demote the false-positive patterns, reconcile against the current prompt…
pr-review
Review open PRs two ways - default is a per-PR deep review with severity-tagged findings, inline comments, and a verdict; --survey runs a risk-tiered triage digest of what's safe to merge first.
pr-triage
First-touch triage for external pull requests - verdict, label, and a welcoming comment within minutes of open.