Skip to content

gemini-code-review.md

Latest commit

 

History

History
69 lines (57 loc) · 4.07 KB

gemini-code-review.md

File metadata and controls

69 lines (57 loc) · 4.07 KB
description Cheap first-pass code review (one-shot via liteLLM) on a parametrizable model — Gemini 3.5 Flash by default, or any provider via --model; supports --adversarial. Curated by the orchestrator against the local repo's CLAUDE.md. Design B — no nested agent.
argument-hint [PR# | branch | --uncommitted] [--model <id>] [--adversarial] (default: <base>...HEAD, gemini-3.5-flash)

Gemini Code Review: $ARGUMENTS

A cost-optimized first-pass code review. The heavy diff-reading runs on Gemini 3.5 Flash (cheap/fast) in a single one-shot request; you (the orchestrator, on a Claude model) then curate the output against the local repo's review rules. This is NOT a replacement for a full Claude-model review on high-risk PRs — it's a fast triage pass.

How it works (Design B)

${CLAUDE_PLUGIN_ROOT}/scripts/gemini-code-review.sh resolves the diff, starts a transient liteLLM proxy, sends the diff + a condensed review rubric to gemini-3.5-flash in ONE completion, and returns review markdown. No nested Claude Code agent runs on Gemini, so there is no tool-call-translation fragility.

Steps

  1. Run the worker against the requested target (default <base>...HEAD, base auto-detected from origin/HEAD → develop/main/master). The worker is multi-model: --model <litellm-id> picks the backend (default gemini/gemini-3.5-flash; also openai/gpt-5.4-mini, anthropic/claude-...). It needs the matching provider key in the environment — gemini/* → GEMINI_API_KEY, openai|gpt* → OPENAI_API_KEY, anthropic|claude-* → ANTHROPIC_API_KEY. Provide it via the secret helpers so it never leaks:

    # one-time, if not already staged:  /secret-input
# default (Gemini):
/secret-use GEMINI_API_KEY -- bash "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-code-review.sh" $ARGUMENTS
# cross-provider second opinion (kills sycophancy — different model, different blind spots):
/secret-use OPENAI_API_KEY -- bash "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-code-review.sh" --model openai/gpt-5.4-mini $ARGUMENTS
# adversarial (devil's advocate — challenges design/assumptions/trade-offs):
/secret-use GEMINI_API_KEY -- bash "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-code-review.sh" --adversarial $ARGUMENTS
    • $ARGUMENTS may be a PR number, branch, --uncommitted, an a..b range, or empty — optionally with --model <id> and/or --adversarial. Pass it through verbatim.
    • If the script exits non-zero, surface the stderr reason (missing provider key, empty diff, proxy failed to start) and STOP — do not fabricate a review.

  2. Curate the Gemini output — do NOT just relay it. For each finding:

    • Validate against the actual diff and this repo's CLAUDE.md (and any nearest-directory CLAUDE.md / docs/agent-standards). Confirm, correct, or drop false positives (Flash over-flags).
    • Add repo-rule-specific issues Flash missed that are visible in the diff (architecture-boundary violations, banned patterns, test conventions, RLS / security rules — whatever the repo's standards encode).
    • Keep file:line precision; discard vague advice.

  3. Present the curated review in the worker's shape (Files table, Critical/Major/Minor findings, Missing Tests, Verdict + metrics) with a one-line header noting it is a Gemini-3.5-Flash first pass curated by the orchestrator.

When to use

  • /make-no-mistakes:gemini-code-review: cheap triage, WIP self-review, large/low-risk diffs, fast feedback.
  • --model openai/gpt-5.4-mini (or any other provider): a genuine cross-provider second opinion — a different model has different blind spots, so it catches what the default misses (same premise as OpenAI's codex-plugin-cc).
  • --adversarial: when you want the design challenged, not approved — surfaces unstated assumptions, rejected trade-offs, and failure modes on load-bearing PRs.
  • A full Claude-model review (e.g. a repo's own /code-review): authoritative review on high-risk PRs (auth, payments, migrations, RLS, infra).