Borrowing it
Nothing to install: this file belongs to Pratiyush/llm-wiki. 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/Pratiyush/llm-wiki/master/.claude/commands/review-pr.mdgit clone --depth 1 https://github.com/Pratiyush/llm-wikiWrote 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/commands/pratiyush/llm-wiki/review-pr)<a href="https://agentmods.dev/commands/pratiyush/llm-wiki/review-pr"><img src="https://agentmods.dev/badge/commands/pratiyush/llm-wiki/review-pr/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/commands/pratiyush/llm-wiki/review-pr"><img src="https://agentmods.dev/badge/commands/pratiyush/llm-wiki/review-pr.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 | $0.00950 |
| Opus 5 | $0.00000 | $0.00475 |
| Sonnet 5 | $0.00000 | $0.00190 |
| Haiku 4.5 | $0.00000 | $0.00095 |
Grade C, and why
review-pr scanned grade C with 1 finding 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- What the contributor should do to get to Approve --> How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run the canonical llmwiki code review against a pull request and post findings.
Usage: /review-pr
This slash command loads docs/maintainers/REVIEW_CHECKLIST.md and
applies every section to the diff of the given PR. The rule is: one
concern per bullet, blocker-vs-nit classification, link to the
relevant doc section for every finding.
Workflow
-
Resolve the PR number — accept either
123orhttps://github.com/Pratiyush/llm-wiki/pull/123. Usegh pr view <N>to fetch the title, body, author, and file list. -
Read the governance docs first — load all of:
docs/maintainers/REVIEW_CHECKLIST.md(the full bar)docs/maintainers/ARCHITECTURE.md(layer boundaries)docs/maintainers/DECLINED.md(prior rejections that may apply)CONTRIBUTING.md(TL;DR rules)SECURITY.md(privacy + security bar)
-
Fetch the diff —
gh pr diff <N>for the raw patch. Walk the diff file by file, mapping each to its layer viadocs/maintainers/ARCHITECTURE.md. -
Run the checklist — apply every section of
REVIEW_CHECKLIST.mdto the diff:- Meta (linked issue, one concern, commit title, CHANGELOG, tests, CI)
- Layer boundaries (layer-appropriate changes, no new runtime deps)
- Security + privacy (no real session data, XSS, no network, localhost binding, no telemetry)
- Code quality (docstrings, comments, error handling, type hints, dead code)
- Tests (happy + edge, descriptive names, regression, tmp_path, run locally)
- Docs (README, CHANGELOG, docs/, docstrings)
- Build + runtime smoke
-
Classify each finding — two categories only:
- Blocker — Meta section failures, security/privacy issues, failing tests, failing build, broken layer boundaries
- Nit — style, wording, missing comment, cosmetic
-
Post the review — group findings by file, then by severity within each file. Blockers first. Every finding cites the relevant section of
REVIEW_CHECKLIST.mdso the contributor can look up the rule. -
Final verdict — one of:
- Approve — no blockers, maybe a few nits
- Request changes — at least one blocker
- Comment — questions only, no blockers, no nits (rare)
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 · 103 lines · 0 tokens per session scan C f5a889c36337
review-pr is a command published in the GitHub repository Pratiyush/llm-wiki (386 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 950 tokens. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
sdd-continue
Continue the next SDD phase in the dependency chain.
sdd-status
Show structured SDD status for an active change.
sdd-new
Start a new SDD change — runs exploration then creates a proposal.
gentle-sdd-ff
Fast-forward all SDD planning phases — proposal through tasks.
sdd-explore
Explore and investigate an idea or feature — reads codebase and compares approaches.
sdd-init
Initialize SDD context — detects project stack and bootstraps persistence backend.