FE-786: Add a skill for "high level" PR review

[lunelson]

No description provided.

[lunelson]

• FE-786: Add a skill for "high level" PR review #161 👈 (View in Graphite)
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[cursor]

PR Summary

Low Risk
Changes are limited to agent skill documentation and test assertions on that doc; no production runtime or auth paths are modified.

Overview
Adds ln-pr-review, an agent skill that guides high-level, architectural PR review (design shift, complexity, precedent, scope) instead of line-level nitpicks. The flow is interactive: load the PR with gh, triage change shape, surface 0–5 concerns with the user, draft a short plain-English comment, then submit a formal gh pr review verdict (approve, comment, or request changes) using --body-file so review text is not truncated by shell heredocs.

Also adds scripts/skill-docs.test.ts, which reads the skill markdown and asserts key safety contracts stay in place: stable <pr> / -R binding, stop when the queue or diff is empty, correct verdict semantics vs gh pr comment, and no heredoc-based review body writes.

^{Reviewed by Cursor Bugbot for commit 0355d22. Bugbot is set up for automated code reviews on this repo. Configure here.}

[augmentcode]

🤖 Augment PR Summary

Summary: Adds a new agent skill for doing “above-the-line” PR reviews focused on architectural/strategic concerns.

Changes:

• Introduces .agents/skills/ln-pr-review/SKILL.md with an interactive workflow for triaging PR “shape” and surfacing 0–5 high-level concerns
• Standardizes a short review flow: load PR context via gh-axi, discuss candidate concerns with the user, then post a concise GitHub review comment
• Provides guidance for choosing verdicts (approve vs request changes) and for keeping feedback plain-English and scoped
• Includes a posting recipe using gh pr review --body-file to avoid shell quoting pitfalls
• Adds a routing table to hand off to follow-up workflows (next PR, issue, scope, done)

_{🤖 Was this summary useful? React with 👍 or 👎}

[augmentcode]

Review completed. 1 suggestion posted.

Comment augment review to trigger a new review at any time.

[cursor]

Cursor Bugbot has reviewed your changes and found 3 potential issues.

There are 7 total unresolved issues (including 4 from previous reviews).

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 81a4a57. Configure here.}

[Copilot]

Pull request overview

Adds a new local agent skill (ln-pr-review) intended to guide “high level” PR reviews (architectural/strategic lens), and introduces a small Vitest contract test to enforce key safety/verdict semantics in the skill doc.

Changes:

• Introduces the .agents/skills/ln-pr-review/SKILL.md skill with an interactive workflow and explicit gh pr review submission guidance.
• Adds scripts/skill-docs.test.ts to assert specific “safety contract” phrases/constraints in the skill doc (e.g., PR identity binding, stop-on-ambiguity, verdict semantics, avoiding heredocs).

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
scripts/skill-docs.test.ts	Adds a Vitest contract test that validates key invariants in the ln-pr-review skill documentation.
.agents/skills/ln-pr-review/SKILL.md	Adds the new “high level PR review” skill guidance, including loading/binding rules and formal gh pr review verdict behaviors.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.