claude-skills

Get a second opinion on your current implementation plan by sending it to Codex for independent review. Use after brainstorming, after writing a plan, or when you want to iterate on a plan before execution. Triggers on 'get a second opinion', 'review plan with codex', 'codex review', or 'iterate on plan'.

Use when reviewing an implementation plan document (docs/superpowers/plans/*.md or similar) for LLM-style overengineering before implementation. Runs a rubric-driven Claude pass plus a Codex adversarial pass, merges findings, and proposes specific edits to slim the plan. Triggers on 'check this plan for bloat', 'is this overengineered', 'detect overengineering', or invoked as /detect-overengineering.

Use when the user wants a git-history health check of an unfamiliar or inherited repository — triggers on "git recon", "codebase health check", "analyze this repo's history", "where are the landmines", or onboarding/due-diligence framing. Do NOT use for architecture review, bug-fixing, or feature work; this skill reads history, not code. Main-thread only — subagents dispatched to execute a delegated task must NOT invoke this skill.

Use when the user needs to create a presentation, pitch deck, slide deck, or visual summary about a project, codebase, website, or topic. Triggers on requests like 'build a deck', 'create a presentation', 'make slides about', or 'pitch deck for'.

Use when reading or reporting what automated PR reviewers left before merging — Codex (`chatgpt-codex-connector`) and an Architecture-Review / SAR GitHub Action (`github-actions[bot]`). Triggers on "what did the bots say", before declaring a PR done or merging, or when a Codex finding or clean verdict seems missing because `gh pr view --json comments` came back empty. Read-only (does not commit or push); for auto-fixing PR feedback use `/pr-fix` instead.