craft-skills

Use when the user needs to plan and architect a new feature before implementation begins. This includes analyzing requirements, asking clarifying questions, and creating a detailed implementation plan aligned with the project's architecture. Invoke BEFORE any coding work starts.

Use when implementation + verification are complete and the work needs to be integrated — commit, open a PR, merge, or clean up the branch. The final step of the craft pipeline (after develop/test/report). Works in the current checkout; no worktree assumption.

Use when starting a new feature that needs thorough design exploration. This is the most comprehensive workflow: collaborative brainstorming with the user, then planning, then parallel agent development, then browser testing. Use for complex features, unclear requirements, or when the user wants to explore approaches before committing.

Post-develop visual regression + design-contract audit. Starts the dev server, captures screenshots of changed UI routes, and dispatches a Haiku-vision agent to compare the result against .claude/aesthetic-direction.md and the feature's ux-brief.md. Flags regressions before browser-test runs. Escalates to Sonnet or Opus only when Haiku flags issues.

Use when an approved implementation plan exists and is ready to be executed. Initializes shared state, dispatches parallel developer agents, performs integration review, and runs full verification (lint, tsc, format, build).

Generate or refresh .claude/reuse-index.md — a one-page inventory of the project's shared utilities, enums, hooks, and components that downstream implementer agents must consult before writing new code. Project-agnostic scan: no framework assumptions baked in. Invoke once per project (or when the shared surface changes meaningfully).

Generate a structured UX brief for a feature before implementation — diagnosis + prioritized patches + success criteria — that downstream implementer agents (Sonnet, Qwen3.6, Haiku) execute without clarification. Invoked by architect/craft when the spec contains UI components. Solo frontend-design by default; combined mode (frontend-design + ui-ux-pro-max) for complex UI based on spec complexity tag.

Reference for code-review-graph MCP queries. Calling skills run the graph tools inline via ToolSearch in the main conversation — no dedicated agents. Covers graph freshness, embedding setup, semantic search, and targeted queries.

Reference for local-LLM scripts. Calling skills run the bash commands inline in the main conversation — no dedicated agents. Covers the availability check, model loading, script execution, and unloading lifecycle.

Loaded at session start. Establishes awareness of craft-skills so the assistant invokes relevant skills before acting. Not invoked directly.

Use when an approved implementation plan already exists (from craft, implement, or architect) and needs to be executed and tested. Combines develop + browser-test into a single pipeline.

Use when auditing, improving, or maintaining project health — CLAUDE.md accuracy, memory hygiene, settings, plans, and pattern discovery. Invoke periodically or when the user asks to audit or improve their project configuration.

Cost-optimized craft variant: Qwen3.6-35B-A3B handles implementation and reviews locally, Opus orchestrates, Sonnet as fallback. Requires LM Studio running. Targets ~45-60% API cost reduction vs base /craft.

Generate or verify the project's one-time AESTHETIC_DIRECTION.md — a design-language contract used by ux-brief and design-review. Invoked automatically by architect/craft when UI work is planned and .claude/aesthetic-direction.md does not yet exist. Auto-generates without blocking; user can refine the file later.

Use after implementation is complete to review changed code for reuse opportunities, quality issues, and architecture compliance. Identifies missed shared components, unnecessary complexity, and boundary violations.

Use when a feature has been built and needs browser-based UI testing. Plans test scenarios, groups them into parallel batches, and dispatches multiple browser-tester agents simultaneously.

Use when investigating bugs, errors, failing tests, or unexpected behavior. Enforces systematic root-cause investigation BEFORE attempting any fix. Invoke this skill before writing any fix code.

Use when implementing a feature with clear, well-understood requirements. Quick planning via architect agent, then parallel agent development, then browser testing. Faster than craft — use when requirements don't need deep exploration.