How do I install the Pr Review skill?

Run npx skillmds@latest add neomjs/pr-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Pr Review safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Pr Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Pr Review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Pr Review?

neomjs (@neomjs) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Pr Review by neomjs — Agent Skill

Name: Pr Review
Author: neomjs

Pr Review

Standardized guidelines and templates for structuring Pull Request reviews so feedback is actionable, encouraging, and extractable by the Native Edge Graph. MANDATORY ROI WARNING: Skipping the review template guarantees CI lint failure. Triggers: Reviewing a PR (yours or peer's) — structured eval metrics, graph ingestion tags, severity ladder, restates §0 merge gate, post-comment A2A commentId hand-off (reviewer→author) per guide §10, Evidence Audit + Source-of-Authority sections (template §) for substrate/runtime-AC PRs and authority-citation review-comments.

neomjs Updated Sep 21, 2026

File contents

PR Review Skill

If you are tasked with conducting a Pull Request review, generating feedback, or helping a user formulate a PR Review, you MUST immediately use the view_file tool to read and strictly adhere to .agents/skills/pr-review/references/pr-review-guide.md before proceeding.

Prior-art sweep gate (PR-review = the last line of defense, per AGENTS.md §verify_before_assert): before scoring, run a cheap 3–10-call query_raw_memories / query_summaries sweep of the PR's decision space — a prior session may have settled the shape, an ADR may already govern it, or the same wrong-shape may have been caught before. The sweep RESULT is the V-B-A; CI-green ≠ AC-met (#13354). Same gate, on your instrument: if the diff adds a capability gate / flag / new field, or your review is about to claim something is absent, load references/reviewer-instrument-audit.md — does satisfying the gate cause an effect or only describe one, does a production writer exist for the field, and does any "no caller" search carry a positive control plus the tree/SHA it ran against.

Templates: Round 1: load pr-review-template.md (full structure) — or, for a mechanical/micro PR per guide §6.4 eligibility, pr-review-micro-review-template.md (the blast-scaled light form). Round 2: load pr-review-round-2-template.md — disposition only, actions verbatim. Budget: ONE ordinary RC per reviewer family; at >24KB load audits/review-cost-circuit-breaker.md. Then approve/polish, A+FU, or one terminal D+S—never another ordinary RC.

---name: pr-reviewdescription: Standardized guidelines and templates for structuring Pull Request reviews so feedback is actionable, encouraging, and extractable by the Native Edge Graph. MANDATORY ROI WARNING: Skipping the review template guarantees CI lint failure. Triggers: Reviewing a PR (yours or peer's) — structured eval metrics, graph ingestion tags, severity ladder, restates §0 merge gate, post-comment A2A commentId hand-off (reviewer→author) per guide §10, Evidence Audit + Source-of-Authority sections (template §) for substrate/runtime-AC PRs and authority-citation review-comments.---# PR Review SkillIf you are tasked with conducting a Pull Request review, generating feedback, or helping a user formulate a PR Review, you MUST immediately use the `view_file` tool to read and strictly adhere to `.agents/skills/pr-review/references/pr-review-guide.md` before proceeding.**Prior-art sweep gate (PR-review = the last line of defense, per AGENTS.md §verify_before_assert):** before scoring, run a cheap 3–10-call `query_raw_memories` / `query_summaries` sweep of the PR's decision space — a prior session may have settled the shape, an ADR may already govern it, or the same wrong-shape may have been caught before. The sweep RESULT is the V-B-A; CI-green ≠ AC-met (#13354). **Same gate, on your instrument:** if the diff adds a capability gate / flag / new field, or your review is about to claim something is absent, load `references/reviewer-instrument-audit.md` — does satisfying the gate cause an effect or only describe one, does a production writer exist for the field, and does any "no caller" search carry a positive control plus the tree/SHA it ran against.**Templates:** **Round 1:** load `pr-review-template.md` (full structure) — or, for a mechanical/micro PR per guide §6.4 eligibility, `pr-review-micro-review-template.md` (the blast-scaled light form). **Round 2:** load `pr-review-round-2-template.md` — disposition only, actions verbatim. **Budget:** ONE ordinary RC per reviewer family; at >24KB load `audits/review-cost-circuit-breaker.md`. Then approve/polish, A+FU, or one terminal D+S—never another ordinary RC.

Frequently asked questions