stefan-stepzero/shipkit Skills

Maps Claude Code changes to Shipkit gaps. Reads scout report and audits all skills, agents, hooks, and settings against current CC capabilities. Identifies deprecated patterns, missing features, and update opportunities. Use after running shipkit-scout.

Two modes: (1) Solution Architect — proposes complete architecture from product definition, stack, and specs. (2) Decision Logger — logs individual architecture decisions. Triggers: 'propose architecture', 'log decision', 'architecture choice'.

Build and compile relentlessly until success — keeps fixing build errors, type errors, and compilation failures in a loop. Use for migrations, setup, major refactors, or when you need to make it compile.

Update CLAUDE.md with learnings, preferences, or project context. Use when user says "update claude.md", "add to claude.md", or wants to persist something.

List and clean up stale implementation worktrees created by /shipkit-implement-independently. Shows PR status and age for each worktree. Use when worktrees accumulate and need cleanup.

Generate project index for faster codebase navigation. Triggers: 'index codebase', 'create index', 'map project'.

Use when user wants to visualize or present shipkit content as HTML. Triggers: 'create presentation', 'visualize this', 'HTML report'.

Breaks a framework spec into implementation steps with file ownership, dependencies, and phase gates. Outputs a plan compatible with shipkit-dev-team for parallel execution. Use after running shipkit-dev-spec.

Manages Shipkit release process — version bump, validation, changelog, counts update, and pre-push checklist. Ensures nothing ships broken. Use when preparing a new Shipkit version for GitHub.

Reviews framework changes for design quality, pattern consistency, CC compatibility, and Skill Value Test compliance. Goes beyond structural integrity to catch design issues. Use after implementing framework changes, before release.

Specs framework changes for the Shipkit repo — new skills, agents, hooks, refactors, or architecture changes. Forces design decisions before implementation. Outputs structured spec to .claude/specs/. Use when planning any non-trivial framework change.

Orchestrates Shipkit framework development using Agent Teams. Creates coordinated teams from dev plans for parallel implementation, review, and release. Use when a framework change is big enough for team execution.

Shows current Agent Team status — active teammates, task progress, phase gates, and blockers. Reads team-state file and task list to produce a progress dashboard. Use during team execution to monitor progress.

Document bugs, ideas, improvements, and decisions. Manages docs/development/manifest.json as the central registry.

Validates Shipkit framework integrity before release. Checks for broken references, manifest/disk sync, installer validity, hook references, cross-skill dependencies, documentation accuracy, and Claude Code changelog compatibility. Auto-fetches latest changelog and flags deprecated patterns, unknown hook events, and feature adoption gaps. Uses caching to skip unchanged files. Use when preparing releases or after refactoring.

Brainstorms new skill opportunities, workflow improvements, and composition patterns based on analyst findings. Produces ranked opportunity cards with effort/impact scores and implementation sketches. Use after running shipkit-analyst.

Fetches latest Claude Code documentation, changelog, and GitHub issues to detect new features, breaking changes, and community patterns. Writes a structured scout report for downstream analysis. Use when checking what's new in Claude Code.