OpenSkillIndex← back to search
claude-code

Claude-skill-registry explore

Systematically explore a codebase area and find/fix obvious issues with fresh eyes.

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/explore-ianwatts22-codex-config" ~/.claude/skills/majiayu000-claude-skill-registry-explore && rm -rf "$T"
manifest: skills/data/explore-ianwatts22-codex-config/SKILL.md
source content

Explore the codebase methodically, not randomly. Goal: understand the relevant execution flow, identify likely issues, and make the smallest correct improvements.

How To Operate

  1. Clarify the target
    • If 
      FOCUS
      is provided, use it. If not, ask what area/bug/goal matters most.
    • Ask for repro steps, expected behavior, and constraints.
  2. Map the flow
    • Identify entrypoints (CLI entry, API route, job, UI page, DB mutation, etc.).
    • Trace the call chain through imports and callers.
    • Note sources of truth (schema, config, shared types).
  3. Look for high-signal problems
    • Obvious bugs, incorrect assumptions, dead code, mismatched types.
    • Inconsistencies with the repo’s conventions (SSOT, validation, config).
    • Missing error handling / logging around external boundaries.
  4. Fix surgically
    • Prefer minimal diffs over refactors.
    • Don’t introduce new abstractions unless they delete more complexity than they add.
  5. Verify
    • Run the smallest targeted test/build/lint available.
  6. Report
    • What you inspected, what you found, what you changed, what remains.

Output Format

  • Summary (1–3 bullets)
  • Findings (bugs / risks / debt)
  • Fixes (what changed + why)
  • Next steps (if any)