Claude Skills by Rune-kit

Detects scope creep by quantifying drift percentage. Auto-triggered by L1 orchestrators when files exceed the original plan. Compares git changes against plan, classifies drift into 4 tiers: ON_TRACK, MINOR_DRIFT, SIGNIFICANT_DRIFT, OUT_OF_CONTROL.

Fast codebase scanner. Use when any skill needs codebase context. Finds files, patterns, dependencies, project structure. Pure read-only — never modifies files.

Environment-aware pre-flight check. Use when starting work in a new environment, switching machines, or when 'works on my machine' bugs surface. Validates OS, runtime versions, installed tools, port availability, env vars, and disk space BEFORE coding starts. Like sentinel but for the environment, not the code.

Automated security gatekeeper. Blocks unsafe code before commit — secret scanning, OWASP top 10, dependency audit, permission checks. A GATE, not a suggestion.

Step-by-step complex reasoning for multi-variable problems. Breaks interconnected decisions into ordered logical steps with bias detection, reversibility classification, and second-order effect tracking.

Universal context persistence across sessions. Auto-saves decisions, conventions, and progress to .rune/ files. Loads state at session start. Use when any skill makes architectural decisions or establishes patterns that must survive session boundaries.

Use when creating new Rune skills, editing existing skills, or verifying skill quality before deployment. Applies TDD discipline to skill authoring — test before write, verify before ship.

Meta-enforcement layer that routes EVERY agent action through the correct skill. MUST check this routing table before ANY response involving code, files, or technical decisions. Default: route to rune:cook for code tasks. Prevents rationalization, enforces check-before-act discipline.

Generate Marp-compatible slide decks from structured JSON schema. Converts context into presentations for tech talks, sprint demos, and tutorials.

Incremental refactorer. Use when refactoring ONE module at a time within a rescue workflow after safeguard has set up safety nets — never as a standalone refactor for greenfield code. Applies proven patterns: Strangler Fig, Branch by Abstraction, Expand-Migrate-Contract.

Multi-agent meta-orchestrator. Use when task spans 5+ files or 3+ modules, or when user says 'parallel', 'split this up', 'do all of these'. Decomposes large tasks into parallel workstreams, assigns to isolated cook instances, coordinates merging.

TDD test writer. Writes failing tests FIRST (red), then verifies they pass after implementation (green). Covers unit, integration, and e2e tests.

Scan market trends, competitor activity, and emerging patterns. Monitors Product Hunt, GitHub Trending, HackerNews, and social platforms.

Universal verification runner. Runs lint, type-check, tests, and build. Use after any code change to verify nothing is broken.

Video content planning. Writes narration scripts, storyboards, shot lists, and asset checklists. Saves plan to marketing/video-plan.md.

Post-deploy monitoring. Use when verifying a deployed app is healthy after deploy or launch — auto-fires from launch/deploy. Checks HTTP status, response times, error detection, and smoke test report.

Git worktree lifecycle management. Use when team needs parallel isolated workspaces for multi-stream development, or when an experiment must not touch the main working tree. Creates isolated workspaces, manages branches, handles cleanup. Called by team and cook.