How this skill is triggered — by the user, by Claude, or both
Slash command
/superjoe:superjoeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Superjoe = a crew of specialized agents for high-quality, minimalist, and secure code.
Superjoe = a crew of specialized agents for high-quality, minimalist, and secure code.
| Task | Use |
|---|---|
| Write minimal, high-quality code / Surgical edits | builderjoe |
| Write or update concise, goal-oriented documentation | docujoe |
| Review code for minimalism, architecture, or performance | inspectorjoe |
| Uncover vulnerabilities and perform security research | secretjoe |
| Complex orchestration / High-level design | Main thread |
Rule of thumb: Use the specialized agents for targeted tasks to maintain high quality and minimalism.
By delegating to specialized "joe" agents, the main thread ensures that every piece of code is written by a minimalist (builderjoe), documented precisely (docujoe), reviewed for architecture and performance (inspectorjoe), and vetted for security (secretjoe). This prevents bloat and security regressions.
What main thread can rely on per agent:
builderjoe
Minimalist implementation, following naming-things guidelines, with 100% test coverage.
docujoe
Concise, imperative documentation in present tense, focused on the user's goals.
inspectorjoe
Bullet-point list of issues with location and one-sentence reasoning, focusing on minimalism and security.
secretjoe
Minimal step-by-step proof (QeD) of a vulnerability and a brief explanation of the exploit.
npx claudepluginhub codingjoe/superjoeProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.