go-to-k

review-pr

Recommend the right reviewer set for a PR from what it touches. Outputs a concrete plan (1 reviewer by default, plus the security reviewer or all three axes when a trigger fires) with ready-to-paste Agent dispatch prompts.

go-to-k 147 12 Updated 1mo ago

Resources

1
GitHub

Install

npx skillscat add go-to-k/cdkd/review-pr

Install via the SkillsCat registry.

SKILL.md

PR Review Recommendation

Decide how much review rigor a PR warrants and surface the dispatch prompts. The
recommendation says what a PR needs AT MINIMUM.

The recommended set is a FLOOR, not a cap, and wall-clock / token cost is never
a reason to come in under it or to stop at it
(AGENTS.md, "Cost is not a
tiebreaker"): when unsure, add the axis. Reviewers are read-only agents that run
in parallel.

This skill never spawns reviewers — it reads PR stats, applies the heuristic, and
prints a recommendation. The main session orchestrator issues the Agent
calls, extending the templates in
references/output-template.md with PR-specific
context.

The per-stage detail lives under references/, and reading a stage's file at
stage entry is mandatory
— the one-line summaries below are routing, not the
procedure.

Steps

  1. Is it your turn? A PUSH is not a round-completion signal — a REPLY is.
    Read references/round-completion.md and
    apply it BEFORE opening a round on a head you have not reviewed.

  2. Read the PR — paths above all, plus loc / fc for context and the
    touched src/ files' recent history. Commands and their traps:
    references/pr-stats.md. The history probe is
    mandatory: it is the one step-3 signal not readable off paths.

  3. Default: ONE reviewer (pr-code-reviewer). A docs-only or test-only
    diff is not discounted below it.

  4. Triggers from the paths list and the src/** size —
    references/bias-factors.md, read at this step
    and AUTHORITATIVE for every list below.

    • Size: the src/** diff exceeds 400 lines OR 8 files → add
      pr-spec-reviewer + pr-test-reviewer (all three axes).
    • Security add-on (pr-security-reviewer, additive to whatever else
      runs): a secret / credential / redaction / masking / sensitive-value
      persistence / process-launch surface, src/provisioning/providers/**, or a
      PR that IS a security fix. A security blocker stops the merge like any
      other.
    • All three axes plus security: a state-schema bump, or a security fix.
  5. Resolve the set: default one reviewer, plus the size axes when the size
    trigger fired, plus pr-security-reviewer when its trigger fired (a 3-axis
    security fix dispatches four reviewers, not three). Reviewers run ONCE, on
    the FINAL sha: a fix round is re-checked by messaging the SAME reviewer with
    the delta (SendMessage), never by a fresh dispatch.

  6. Render the recommendation —
    references/output-template.md.

  7. Dispatch reviewers and synthesize —
    references/dispatch-and-marker.md. NEVER
    report a round closed without dispatching the reviewers first.