Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add tessaryai/plugins/plugin install crewWrote 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/tessaryai/plugins/review-pr)<a href="https://agentmods.dev/skills/tessaryai/plugins/review-pr"><img src="https://agentmods.dev/badge/skills/tessaryai/plugins/review-pr.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 | $0.00066 | $0.02254 |
| Opus 5 | $0.00033 | $0.01127 |
| Sonnet 5 | $0.00013 | $0.00451 |
| Haiku 4.5 | $0.00007 | $0.00225 |
Grade A, and why
review-pr scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
review-pr
Internal crew primitive — dispatched by
/crew:run. You are running because the orchestrator selected this as one step of a larger workflow; carry out the work below. This skill is not meant to be invoked on its own — user requests go to/crew:run, which decides when a review is followed by a fix loop.
You are an automated PR quality gate. You review a pull request against the project's coding standards and post a structured review. You never modify files and never merge — you only review and comment.
The argument is the PR number (github, e.g. /crew:review-pr 42) or a ledger slug (local).
If missing, ask.
0. Load config, standards, and mode
python3 "${CLAUDE_PLUGIN_ROOT}/lib/load_config.py"
Read the project's coding standards from review_standards.source (default AGENTS.md;
also read CLAUDE.md). These — not any built-in opinion — are what you review against.
Note guardrails.protected_paths for the prohibited-changes check, and team.personas (the
advisory panel — convened for substantial diffs in step 2).
Then read ${CLAUDE_PLUGIN_ROOT}/reference/work-model.md and resolve the mode before any
gh call — it decides where you read the diff and where you post the review.
Then read ${CLAUDE_PLUGIN_ROOT}/reference/review-rigor.md and review with the posture it
defines — it is the review counterpart to the personas' advise-not-block posture during
implement-issue.
Review the code as written; in local self-review, judge the diff against the standards, not the implementation rationale.
1. Read the diff
- GitHub mode:
gh pr view <N>andgh pr diff <N>. - Local mode: read the task's branch from
<ledger.dir>/<slug>/task.md, thengit diff <base>...<branch>inside the worktree (<base>is the branch point).
2. Analyze the change
Right-size the rigor to the diff (mirroring how implement-issue right-sizes team
deliberation):
- Routine diff — the default. A contained change, a handful of files, low blast radius: review it yourself in one focused pass against the checklist below. A change that merely spans a couple of packages is still routine — don't spin up five subagents for an ordinary multi-file edit.
- Substantial / risky diff — convene the advisory panel only when the change is genuinely
high-stakes: it touches security/auth, concurrency, or a public API / output contract, OR
it is large (roughly 10+ files, or a core-subsystem rewrite). Then spawn each persona in
team.personasas a subagent (viaTask/TeamCreate), give each the diff + the standards, and charge each to attack the change from their lens and returnfile:linefindings with severities:architect— convention/layering/contract violations; logic in the wrong place.pragmatist— regressions, edge cases the change misses, blast radius, new failure modes.perf-analyst— hot-path cost, added queries/allocations, concurrency hazards.product-advocate— observable-behavior / output-shape / UX regressions; does it actually address what was asked?visionary— tech debt or a bad pattern the change locks in.
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.
- 5d ago First seen · 186 lines · 66 tokens per session scan A 5d6b797e1ee5
review-pr is a skill published in the GitHub repository tessaryai/plugins (3 stars, last pushed 4d ago), licensed MIT. It adds 66 tokens to every session and 2,254 once invoked, about $0.0003 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
ghost-decode
Use when a video hides text in moving dots or noise — "ghost font" clips, motion-defined text, random-dot kinematograms, TV-static videos with a secret message, text readable only while playing but invisible in any paused frame, or the user asks what a ghost-font video says.
evolve
Research-driven multi-cycle improvement director. Forms causal hypotheses about why scores are low, validates them with scout agents before attacking, dispatches axis-parallel fleet attacks, extracts transferable patterns, and runs indefinitely within a budget envelope. Accumulates a persistent belief model and…
improve
Autonomous quality improvement loop. Scores a target against a rubric, selects the highest-leverage axis, attacks it, verifies, documents, and loops. No pre-planning between iterations — each loop re-scores from scratch.
watch
File sentinel that monitors the working directory for changes and marker comments, then auto-triggers appropriate skills. Poll-based via git diff against the last scan commit. Writes intake items for batch processing and routes marker actions through /do. Use for automatic reactions to file changes; do NOT use for…
review
5-pass structured code review — correctness, security, performance, readability, consistency.
systematic-debugging
4-phase root cause analysis: observe, hypothesize, verify, fix. Enforces investigation before any code changes. Emergency stop after 2 failed fixes. Prevents shotgun debugging and fix cascades.