rcrdk
- 9 skills
- 0 followers
- 5 hours ago last updated
- ▌
- ▌ Clean · rcrdkActivated with /clean. Passes over generated code and removes imprecision, vagueness, and typical AI patterns — vague code, generic names, obvious comments, premature abstractions, and cosmetic error handling. Use after heavy code generation or when the code "looks right but is not precise".
- ▌ Debugger · rcrdkActivated with /debug. Investigates bugs, finds root cause, proposes a minimal fix, and creates a regression test. Use for runtime errors, stack traces, incorrect behavior, test failures, or intermittent bugs.
- ▌ Interviewer · rcrdkSkill (list in project.mdc or cite in the request). Socratic interview to clarify requirements before implementing. Use before large features, when the requirement is vague, there are multiple interpretations, or the cost of building the wrong thing is high.
- ▌ Performance · rcrdkActivated with /perf. Identifies performance bottlenecks, defines a baseline, separates by category (CPU, I/O, database, network, bundle), and suggests optimizations by impact vs. effort. Use for slow endpoints, heavy queries, slow screens, high memory use, or large bundles.
- ▌ Code Reviewer · rcrdkActivated with /review. Critical code review covering bugs, business logic, edge cases, performance, and patterns. Use after implementing any feature, bug fix, or refactor.
- ▌ Logic Reviewer · rcrdkActivated with /logic. Reviews business rules, conditional flows, calculations, states, and edge cases. Use when unsure whether the code correctly implements the requirement, for critical calculations, state machines, complex conditional flows, or inconsistency between what was asked and what was built.
- ▌ Evidence Verifier · rcrdkProof checklist for staging/prod before asking for human retest. Use after a fix or when project.mdc has a Verification section.
- ▌ Security Reviewer · rcrdkActivated with /secure. Reviews attack surfaces, classifies risk with severity, and proposes immediate and structural fixes. Use for authentication, authorization, APIs, uploads, secrets, input validation, and sensitive data access.