Skip to content

urmzd/dotfiles / dot_agents/skills/review-diff

review-diff

Verified

Standardized review of the CURRENT working-tree changes -- staged, unstaged, and untracked -- against an explicit five-dimension rubric (correctness, security, tests, scope-creep, conventions). Emits a prioritized findings list, each with file:line, severity, and a suggested fix. Read-only: it reviews and reports, it does not commit, push, or edit. Use when the user says "review my changes", "review the diff", "look over my staged changes before I commit", or "what's wrong with this PR locally". Do NOT use for red CI/pipeline failures (use diagnose-ci) or whole-repo quality-framework questions (use assess-quality); this reviews the working-tree diff only. assess-quality and review-design both EXCLUDE day-to-day review and name this skill as its home.

FileKindSize
SKILL.mdskill md5.6 KB
evals/eval_queries.jsonconfig3.2 KB

evals/eval_queries.json

View on GitHub
{
  "skill": "review-diff",
  "description": "Routing evals: does the query trigger a working-tree diff review against the rubric? Near-misses route to assess-quality (set the quality bar), review-design (design tradeoffs), diagnose-ci (red pipeline), and ship/pr (commit + push).",
  "queries": [
    {
      "query": "review my changes before I commit",
      "should_trigger": true,
      "note": "Core ask: review staged + unstaged + untracked against the rubric."
    },
    {
      "query": "review the diff and tell me what's wrong",
      "should_trigger": true,
      "note": "Direct diff review with prioritized findings."
    },
    {
      "query": "look over my staged changes and flag anything risky",
      "should_trigger": true,
      "note": "Staged-only scope; security + correctness dimensions lead."
    },
    {
      "query": "do a code review of what I have locally before I open the PR",
      "should_trigger": true,
      "note": "Pre-PR working-tree review, not a remote PR review."
    },
    {
      "query": "are there any bugs or security issues in my current changes",
      "should_trigger": true,
      "note": "Correctness + security dimensions of the rubric on the local diff."
    },
    {
      "query": "check whether my diff has any scope creep or unrelated edits",
      "should_trigger": true,
      "note": "Scope-creep dimension explicitly."
    },
    {
      "query": "did I forget to add tests for the new branch I just wrote",
      "should_trigger": true,
      "note": "Tests dimension on uncommitted work."
    },
    {
      "query": "review my uncommitted work against main and prioritize the findings",
      "should_trigger": true,
      "note": "Base-ref scope plus prioritized findings list."
    },
    {
      "query": "go over these untracked files I just created and see if they're sound",
      "should_trigger": true,
      "note": "Untracked files are in scope; read them in full."
    },
    {
      "query": "set the quality bar for this project and define what good looks like",
      "should_trigger": false,
      "note": "Framework / quality-bar definition -- assess-quality, not a single-diff review."
    },
    {
      "query": "what are the five questions every change should be judged against",
      "should_trigger": false,
      "note": "The quality framework itself lives in assess-quality."
    },
    {
      "query": "help me reason about whether this design decision is reversible and DRY",
      "should_trigger": false,
      "note": "Design tradeoffs / the 'why' layer -- review-design."
    },
    {
      "query": "why did the CI pipeline fail on this branch",
      "should_trigger": false,
      "note": "Red remote pipeline -- diagnose-ci."
    },
    {
      "query": "commit these changes and push them",
      "should_trigger": false,
      "note": "Commit + push is ship, not review."
    },
    {
      "query": "open a pull request with a summary of my commits",
      "should_trigger": false,
      "note": "PR creation -- the pr skill, not a working-tree review."
    },
    {
      "query": "review pull request #42 on GitHub and leave comments",
      "should_trigger": false,
      "note": "Remote PR review with inline comments is out of scope; this reviews the local working tree."
    }
  ]
}
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 470kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 470kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 470kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall