Command
Use when you need to audit a CLAUDE.md file against Anthropic's best practices. Do NOT use to apply changes — use /claude-md-toolkit:optimize-claude-md for that.
Command
Use when you need to audit a CLAUDE.md file against Anthropic's best practices. Do NOT use to apply changes — use /claude-md-toolkit:optimize-claude-md for that.
Command
Use when creating a new CLAUDE.md or bootstrapping one for a project that doesn't have one yet. Do NOT use to improve an existing CLAUDE.md — use /claude-md-toolkit:optimize-claude-md instead.
Command
Use when you want to apply best-practice improvements to an existing CLAUDE.md file. Do NOT use to create one from scratch — use /claude-md-toolkit:init instead.
Command
Use when reviewing branch changes for merge readiness with severity classification. Do NOT use for staged-only review (use /code-review:staged) or PR feedback triage (use /code-review:triage).
Command
Use to prevent CI coverage failures before pushing by checking coverage against CI-configured thresholds. Do NOT use as a replacement for running tests (use your test command directly).
Command
Use when PR comments are false positives that should be closed with justification. Do NOT use for valid findings that need fixes — use /code-review:implement-fix instead.
Command
Use when a code review session has produced findings that need tracking before implementation. Do NOT use without prior review output — run /code-review:staged or :branch first.
Command
Use when REVIEWFIXES.md exists and you want to implement tracked issues. Do NOT use without a REVIEWFIXES.md — run /code-review:fixes-plan first.
Command
Use to verify and mark issues as fixed in REVIEWFIXES.md. Do NOT use without an existing REVIEWFIXES.md — run /code-review:fixes-plan first.
Command
Use when you want to resolve all PR feedback autonomously in a single pass. Do NOT use for draft PRs or when you need manual control over individual fixes.
Command
Use when you need a thorough code review of a pull request. Do NOT use for staged or branch-only reviews (use /code-review:staged or :branch).
Command
Use when pasting review output from /code-review:staged or :branch. Do NOT use for PR triage (/code-review:triage) or to run a new review.
Command
Use after marking issues as fixed in REVIEWFIXES.md to close corresponding GitHub threads. Do NOT use before verifying fixes — use /code-review:mark-fixed first.
Command
Use when you want to review and fix staged changes in a single session before committing. Do NOT use for PR reviews (use /code-review:pipeline) or read-only reviews (use /code-review:staged).
Command
Use to review staged changes before committing. Do NOT use for branch-wide review (use /code-review:branch) or full PR review (use /code-review:pr).
Command
Use to analyze staged changes or branch diff for technical debt and maintainability issues before merge. Do NOT use for bug detection or security review.
Command
Use to triage PR feedback from AI reviewers and verify validity before implementing. Do NOT use without an open PR with bot comments — use /code-review:branch for local review.
Command
Use when facing any error or unexpected behavior that requires diagnosis before fixing. Do NOT use for quick classification only — use /debug:triage for that.
Command
Use to manage an incident through its lifecycle: triage on detection, status updates during, blameless postmortem after. Do NOT use for ordinary debugging — use /debug:debug for single-engineer bug-hunting.
Command
Use when you need multiple alternative approaches before committing to a solution. Do NOT use when a clear solution exists (plan mode) or for comparing known options (/discuss:tradeoffs).
Command
Use to stress-test a proposal by hearing the strongest counter-arguments. Do NOT use for neutral analysis (/discuss:feature) or comparing options (/discuss:tradeoffs).
Command
Use before plan mode to critically analyze a feature proposal for gaps, edge cases, and risks. Do NOT use for simple Q&A, implementation, or code review (/code-review:).
Command
Use when you have 2-4 specific options and need a structured comparison. Do NOT use for open-ended exploration (/discuss:brainstorm) or single-option analysis (/discuss:feature).