This repository also configures its own agents. See what neatcodeai tells them →
Skill Claude CodeCodex
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
Skill Claude CodeCodex
Use when the user expresses a rule (must / never / always / going forward / no longer / forbid / require / cannot / shouldn't) — captures one rule at a time. ALSO use proactively on a fresh repo with empty skills.yaml: call derivecandidates to mine the existing graph for candidate rules, walk through them with the…
Skill Claude CodeCodex
Use after completing a task or before merging — runs a structured review pass against captured rules, planned changes, and recent commits. Output is categorized (Critical / Important / Minor / Strengths) and ALWAYS persisted to .neatcode/reviews/ - .md so the report survives the session.
Skill Claude CodeCodex
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
Skill Claude CodeCodex
Use for any structural question about the codebase — who calls X, what extends Y, where is Z used, what handles route foo. Routes through the graph instead of grep or file reads.
Skill Claude CodeCodex
Use after answering a non-trivial structural question, after shipping placeholder/stub code during autonomous work, or after making a load-bearing decision not specified in the prompt. Proposes filing the answer as a memory note so the next session doesn't re-derive it.
Skill Claude CodeCodex
Use when implementing any feature or bugfix, before writing implementation code.
Skill Claude CodeCodex
Use at logical-unit boundaries — finished a feature slice, completed a TDD cycle, before claiming a task done, before committing. NOT before every file write. Validates the planned changes against the repo's captured rules and surfaces violations with resolution options.
Skill Claude CodeCodex
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always.
Skill Claude CodeCodex
Use when you have a spec or requirements for a multi-step task, before touching code.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: