Exécutez n'importe quel Skill dans Manus
en un clic

Exécutez n'importe quel Skill dans Manus en un clic

mav-bp-code-review

Étoiles10

Forks2

Mis à jour2 juillet 2026 à 21:42

Code review conventions for all projects. Covers mandatory review requirements, review scope, PR sizing, reviewer and author responsibilities, and automated review integration. Applied when creating or reviewing pull requests.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

thermiteau

thermiteau/maverick

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Métiers associésSOC

Basé sur la classification professionnelle SOC

Analystes en assurance qualité des logiciels et testeursProfessions informatiques et mathématiques·SOC 15-1253

SKILL.md

readonly

name	mav-bp-code-review
description	Code review conventions for all projects. Covers mandatory review requirements, review scope, PR sizing, reviewer and author responsibilities, and automated review integration. Applied when creating or reviewing pull requests.
user-invocable	false
disable-model-invocation	true

Code Review Standards

Ensure all code changes are reviewed before merging; in Maverick the reviewer is agent-code-reviewer running in-session, backed by branch-protection gates.

Maverick's Rules

No unreviewed merges: at least one approval on protected branches; no self-approval; stale approvals are dismissed when new commits are pushed; all CI checks must pass. Never disable or skip a check to merge faster — fix the check or the code.
Security is out of scope for PR review — it runs as a separate pre-push gate via do-cybersecurity-review (update mode against the diff and impact set) before the PR is opened. By review time, security findings are already surfaced and folded into the PR body.
Review scope: correctness, maintainability, test coverage, documentation. Style and formatting are the linter's job — do not comment on them. Do not review generated code (codegen, lockfiles) line by line, and do not request changes to correct-but-different approaches on personal preference.
Verdict discipline: distinguish blocking findings from non-blocking (prefix "nit:" or "optional:"); demonstrate understanding of the change rather than rubber-stamping; approve once the code is correct, tested, and maintainable — do not hold changes hostage for perfection. AI review findings require judgement — dismiss false positives with reasoning, never silently.
Automated checks pass before review begins — do not review code that fails lint, type checks, or tests.

PR Size

Size	Lines changed	Review quality	Recommendation
Small	< 100	Excellent	Ideal. Aim for this size.
Medium	100 - 400	Good	Acceptable for most feature work.
Large	400 - 800	Declining	Split if possible. Flag to reviewers.
Too large	> 800	Poor	Must split. Reviewers cannot effectively review this.

Split strategies: by layer (backend vs frontend), by feature slice (one endpoint/component per PR), by phase (refactor PR before feature PR), or by dependency (extract shared utilities/models into a preparatory PR).

Detecting Code Review Anti-Patterns

Pattern	Issue	Fix
PR merged without any review	No review gate	Enable branch protection requiring approvals
PR with 1000+ lines changed	Too large to review effectively	Split into smaller, focused PRs
All review comments are style nits	Wasted review effort	Configure linters to catch style; focus on logic
Reviewer approves without comments on large PR	Rubber-stamp review	Reviewers must demonstrate understanding of the change
Author pushes "fix review" commits without context	Lost review trail	Explain what was changed in response to which comment
Self-approval on protected branch	Missing review gate	Configure branch protection to disallow self-approval
Skipping CI to merge faster	Bypassing quality gates	CI must pass; fix failures, do not skip them

Plus depuis ce dépôt

même dépôt

do-adopt

thermiteau/maverick

Scan a project for missing best-practice areas and implement the top recommendation for each gap (linting, unit testing) — installs tools, writes configs, verifies, and commits. Pass 'recommend' to stop after writing recommendations without implementing (replaces the old do-recommend skill).

2026-07-0210

do-code

thermiteau/maverick

Implement a focused code change. Use this skill as the wrapper for any implementation work so the Maverick workflow report captures what was done and so the agent applies the project's coding standards before editing. Intended to be invoked once per task from inside a do-issue-* or do-epic phase, not standalone.

2026-07-0210

do-cybersecurity-review

thermiteau/maverick

Audit a codebase for security risks in one of two modes. In full-audit mode it scans the entire codebase and writes a findings report to docs/security-audit.md (run as part of do-init or on demand). In update mode it reviews only a diff plus the code it could impact, returning a structured findings list as a pre-push gate for do-issue-solo and do-issue-guided. Covers secrets exposure, dependency vulnerabilities, authentication and authorisation patterns, input validation, transport security, and common OWASP risks.

2026-07-0210

do-docs

thermiteau/maverick

Create, restructure, or update technical documentation. Handles greenfield projects, refactoring non-compliant docs, and incremental updates after code changes.

2026-07-0210

do-epic

thermiteau/maverick

Work on a multi-story GitHub epic end-to-end. Builds a DAG from the child stories, groups them into waves, runs waves in parallel via per-story worktrees, ejects PRs that fail agent-code-review for human handling, and propagates blocks to downstream stories. Requires git worktrees.

2026-07-0210

do-init

thermiteau/maverick

Initialise a project for use with Maverick — verifies the GitHub App, installs the CLI if needed, writes the project config with integration tracking, scaffolds docs, generates project skills, runs an initial cybersecurity audit, then commits the changes and opens a PR.

2026-07-0210

name	mav-bp-code-review
description	Code review conventions for all projects. Covers mandatory review requirements, review scope, PR sizing, reviewer and author responsibilities, and automated review integration. Applied when creating or reviewing pull requests.
user-invocable	false
disable-model-invocation	true

Code Review Standards

Ensure all code changes are reviewed before merging; in Maverick the reviewer is agent-code-reviewer running in-session, backed by branch-protection gates.

Maverick's Rules

No unreviewed merges: at least one approval on protected branches; no self-approval; stale approvals are dismissed when new commits are pushed; all CI checks must pass. Never disable or skip a check to merge faster — fix the check or the code.
Security is out of scope for PR review — it runs as a separate pre-push gate via do-cybersecurity-review (update mode against the diff and impact set) before the PR is opened. By review time, security findings are already surfaced and folded into the PR body.
Review scope: correctness, maintainability, test coverage, documentation. Style and formatting are the linter's job — do not comment on them. Do not review generated code (codegen, lockfiles) line by line, and do not request changes to correct-but-different approaches on personal preference.
Verdict discipline: distinguish blocking findings from non-blocking (prefix "nit:" or "optional:"); demonstrate understanding of the change rather than rubber-stamping; approve once the code is correct, tested, and maintainable — do not hold changes hostage for perfection. AI review findings require judgement — dismiss false positives with reasoning, never silently.
Automated checks pass before review begins — do not review code that fails lint, type checks, or tests.

PR Size

Size	Lines changed	Review quality	Recommendation
Small	< 100	Excellent	Ideal. Aim for this size.
Medium	100 - 400	Good	Acceptable for most feature work.
Large	400 - 800	Declining	Split if possible. Flag to reviewers.
Too large	> 800	Poor	Must split. Reviewers cannot effectively review this.

Detecting Code Review Anti-Patterns

Pattern	Issue	Fix
PR merged without any review	No review gate	Enable branch protection requiring approvals
PR with 1000+ lines changed	Too large to review effectively	Split into smaller, focused PRs
All review comments are style nits	Wasted review effort	Configure linters to catch style; focus on logic
Reviewer approves without comments on large PR	Rubber-stamp review	Reviewers must demonstrate understanding of the change
Author pushes "fix review" commits without context	Lost review trail	Explain what was changed in response to which comment
Self-approval on protected branch	Missing review gate	Configure branch protection to disallow self-approval
Skipping CI to merge faster	Bypassing quality gates	CI must pass; fix failures, do not skip them