georgesmomo

analyze-requirements

Turn a need into testable acceptance criteria in Given/When/Then, with edge cases enumerated.

brainstorm

Frame a need — problem, users, scope, risks, success metric — before committing to a spec.

clarify

When a request is ambiguous, act as an analyst — reflect it back and ask one targeted question at a time (each with a recommendation) until the need is crisp, then execute.

Code & Development Listed

code-review

Review a deliverable against its requirements, findings graded by severity.

generate-agent

Turn a description (or an auto-analysis) into a new agent persona, grounded in real named methods and matching the framework's gold-standard shape.

generate-dashboard

Turn a description (or an auto-analysis) into a new custom dashboard page, as a declarative block spec that automatically matches every design the dashboard ships.

generate-skill

Turn a description (or an auto-analysis) into a new skill file, grounded in a real, cited domain standard and matching the framework's gold-standard shape.

implement

Implement a plan task as small, conventional commits — red-green when a test exists, boy-scout cleanup on touched code, no scope creep.

propose-customizations

The "Auto" mode — analyze the project and propose concrete, justified dashboard/skill/agent candidates instead of requiring a description upfront.

Code & Development Listed

security-review

Review a change for secrets, authorization and attack surface against OWASP ASVS + Top 10.

write-adr

Record an architecturally significant decision as a short, durable ADR.

Testing & QA Listed

write-e2e-tests

Author durable, CI-runnable Playwright end-to-end tests — run headed, directly in the browser, by default; falls back to MCP, native browser tooling or headless, always telling the user why.