Claude Skills by troykelly

Use after code review - implement ALL findings; any finding not fixed MUST have tracking issue created; no finding disappears without trace

Use when starting any work session - establishes autonomous operation mode, overriding token limits and time pressure to work until goal is achieved

Use when user requests autonomous operation across multiple issues. Orchestrates parallel workers using Task tool, monitors with TaskOutput, handles SLEEP/WAKE cycles, and works until scope is complete without user intervention.

Use after creating PR - monitor CI pipeline, resolve failures, address review feedback, and merge when all gates pass

Use when committing code - ensures atomic, descriptive commits that leave the codebase in a merge-ready state at every point

Use after implementing features - 7-criteria code review with MANDATORY artifact posting to GitHub issue; blocks PR creation until complete

Use when encountering merge conflicts - handle conflicts cleanly, verify resolution, and maintain code integrity

MANDATORY when designing schemas, writing migrations, creating indexes, or making architectural database decisions - enforces PostgreSQL 18 best practices including AIO, UUIDv7, temporal constraints, and modern indexing strategies

Use when encountering failures - assess severity, preserve evidence, execute rollback decision tree, and verify post-recovery state

Use when user-facing features change. Ensures features documentation is updated. Pauses work if documentation has drifted, triggering documentation-audit skill.

Use when receiving UAT feedback, bug reports, user testing results, stakeholder feedback, QA findings, or any batch of issues to investigate. Investigates each item BEFORE creating issues, classifies by type and priority, creates well-formed GitHub issues with proper project board integration.

MANDATORY before any GitHub project operations - caches project metadata to prevent rate limit exhaustion. Called by session-start. Other skills MUST use cached data.

IPv6 is THE first-class citizen. All code, tests, documentation, and configurations MUST be IPv6-first. IPv4 MAY be added only for legacy support as a second-class citizen.

Use when an issue is too large for a single task - breaks into linked sub-issues with full documentation, ensuring manageable work units

Use when code changes touch database, cache, queue, or other service-dependent components - enforces testing against real local services instead of mocks

Use when asked for an architecture blueprint or explicitly asked to run the code-architect subagent.

Use when asked to trace existing codepaths or explicitly asked to run the code-explorer subagent.

Use when explicitly asked to run the code-reviewer subagent or when another skill requires the code-reviewer agent card.

Use when asked to simplify recently changed code without changing behavior or explicitly asked to run the code-simplifier subagent.

Use when asked to review code comments for accuracy/quality or explicitly asked to run the comment-analyzer subagent.

Use when asked to evaluate PR test coverage or explicitly asked to run the pr-test-analyzer subagent.

Use when explicitly asked to run the security-reviewer subagent or when another skill requires the security-reviewer agent card.

Use when asked to detect silent failures/weak error handling or explicitly asked to run the silent-failure-hunter subagent.

Use when asked to evaluate type design/invariants or explicitly asked to run the type-design-analyzer subagent.