implement
Use when ready to build from a spec, design, or plan, or to resume from notes. Triggers include "implement this", "build this", "implement the spec/design/plan", "continue implementation", and "resume where we left off".
What this skill does
# Implement
Act as orchestrator. Dispatch work to sub-agents via the Task tool and record what happens. Do not write code, run tests, or review code directly. Every implementation, testing, and review action goes through a Task tool invocation. No exceptions.
## Input
Invoked as `/implement <path>` where `<path>` is a lore artifact: spec, design, plan, or notes file. Read the input artifact and any lore documents it references. If the input is a notes file, resume from where it left off.
## Output
Implemented code plus a notes file at `.lore/work/notes/<artifact-name>.html`. Load `${CLAUDE_PLUGIN_ROOT}/shared/document-schema.md` for the meta tag fields before writing.
The notes file needs enough structure to be resumable: a progress tracker (phases with checkboxes) and a log (what happened, failures, decisions, discoveries). Update it after every phase, not just at session end.
The output is HTML — make the progress tracker visual. Phases as a checklist with status chips (pending / in progress / done / failed), a collapsible log per phase, and a summary banner at the top showing overall state. Someone opening this mid-session should be able to read the situation in five seconds. Inline CSS and JS are fine; no external dependencies.
## Process
### 1. Initialize
Search for related prior work: invoke the `lore-researcher` agent via Task tool with the artifact description. Wait for the result before continuing.
Break the input into implementable phases. If the input is a plan, phases are its steps. If a spec or design, break into independently testable chunks.
**Task file detection.** When the input is a plan, check whether `.lore/work/tasks/<plan-name>/` exists. If it does, read task files sorted by their `sequence` meta field — these become the phases. Compare the plan's modification timestamp against the oldest task file. If the plan is newer, warn the user and offer three options: re-run `/plan-breakdown`, use existing tasks, or abort.
**Select agents.** Consult `.lore/lore-agents.md` if it exists. Match agents to three mandatory roles:
| Role | Registry Category | Fallback |
|------|-------------------|----------|
| Implementation | Implementation | `general-purpose` |
| Testing | Testing | `general-purpose` |
| Review | Code Quality | `general-purpose` |
### 2. Execute Phases
For each phase:
**a. Implement.** Dispatch to an implementation agent via Task tool. Include: what to build, relevant file paths, context from prior phases or failures. One phase at a time — the agent does not see the full plan.
**b. Test.** Dispatch to a testing agent via Task tool. Include: which files changed, what behavior to verify, how to run the test suite. Expect back: pass/fail and notable findings only.
**c. Review.** Dispatch to a review agent via Task tool. Include: which files to review, relevant requirements from the source artifact. Expect back: non-conformances only.
**d. Handle failures.** Route test or review failures back to an implementation agent via Task tool for correction. Re-dispatch only the failing step, not the full cycle. After two consecutive failed attempts on the same issue, escalate to the user.
**e. Record.** Update the notes file: mark the phase complete, log anything worth preserving (failures, unexpected behavior, decisions not specified in the source).
When phases come from task files, update the task file's `status` meta to `complete` after the cycle passes. If the implementation agent reports work is already done (task file still says `pending`), surface this to the user before skipping.
### 3. Validate
After all phases complete, dispatch a review agent with the full source artifact. Directive: validate the implementation against the source, flag requirements not met or behavior that diverges. This is a holistic check, not a code quality review.
Route any findings back through the implement/test/review cycle.
### 4. Finalize
Update the notes file status to `complete`. Summarize: what was built, how many phases, any divergences.
Suggest: `Run /simplify on the changed files to clean up for clarity.`
## Escalation
Two conditions require human intervention. Everything else is autonomous.
1. **Stuck loop**: Two consecutive failed attempts on the same issue. Present the failure history.
2. **Plan divergence**: Implementation requires something the source artifact didn't specify or contradicts. Present the divergence and ask the user to authorize or redirect.
Do not ask for confirmation between phases.
## Divergence
If reality requires something the source artifact didn't account for, do not proceed. Escalate to the user with the specific divergence and why it's needed. Record approved divergences in the notes file.
Related in Design
contribute
IncludedLocal-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers at ~/.contribute-system/research/, append-only event log at ~/.contribute-system/log.jsonl. No database, no cloud calls. Use when the user asks about their PRs / issues / contributions, wants to find new work to take on, claim an issue, build/refresh a repo's dossier, or draft a Design Issue or PR. Trigger with "/contribute", "what's my PR status", "find a contribution", "claim issue X", "draft a Design Issue for Y", "refresh dossier for Z".
architectural-analysis
IncludedUser-triggered deep architectural analysis of a codebase or scoped subtree across eight modes — information architecture, data flow, integration points, UI surfaces, interaction patterns, data model, control flow, and failure modes. This skill should be used when the user asks to "diagram this codebase," "map the architecture," "show the data flow," "give me an ERD," "trace control flow," "find the integration points," "verify the layout pattern," "audit the UX architecture," or any similar request whose primary deliverable is mermaid diagrams plus cited reports under docs/architecture/. Dispatches haiku/sonnet sub-agents in parallel for per-mode exploration, then verifies every citation mechanically before any node lands in a diagram. Not for one-off prose explanations of code (use code-explanation) or for high-level system design from scratch (use system-design).
mcp
IncludedModel Context Protocol (MCP) server development and tool management. Languages: Python, TypeScript. Capabilities: build MCP servers, integrate external APIs, discover/execute MCP tools, manage multi-server configs, design agent-centric tools. Actions: create, build, integrate, discover, execute, configure MCP servers/tools. Keywords: MCP, Model Context Protocol, MCP server, MCP tool, stdio transport, SSE transport, tool discovery, resource provider, prompt template, external API integration, Gemini CLI MCP, Claude MCP, agent tools, tool execution, server config. Use when: building MCP servers, integrating external APIs as MCP tools, discovering available MCP tools, executing MCP capabilities, configuring multi-server setups, designing tools for AI agents.
react-native-skia
IncludedDesign, build, debug, and optimise high-polish animated graphics in React Native or Expo using @shopify/react-native-skia, Reanimated, and Gesture Handler. Use when the user wants canvas-driven UI, shaders, paths, rich text, image filters, sprite fields, Skottie, video frames, snapshots, web CanvasKit setup, or performance tuning for custom motion-heavy elements such as loaders, hero art, cards, charts, progress indicators, particle systems, or gesture-driven surfaces. Also use when the user asks for fluid, glow, glass, blob, parallax, 60fps/120fps, or GPU-friendly animated effects in React Native, even if they do not explicitly say "Skia". Do not use for ordinary form/layout work with standard views.
plaid
IncludedProduct Led AI Development — guides founders from idea to launched product. Six capabilities: Idea (discover a product idea), Validate (pressure-test the idea against fatal flaws, problem reality, competition, and 2-week MVP feasibility), Plan (vision intake + document generation), Design (translate image references into a design.md spec), Launch (go-to-market strategy), and Build (roadmap execution). Use when someone says "PLAID", "plaid idea", "help me find an idea", "product idea", "idea from my business", "idea from my expertise", "plaid validate", "validate my idea", "pressure-test", "is this idea good", "find fatal flaws", "validate the problem", "plan a product", "define my vision", "generate a PRD", "product strategy", "plaid design", "design from image", "translate image to design", "create design.md", "extract design tokens", "plaid launch", "go-to-market", "launch plan", "GTM strategy", "launch playbook", "plaid build", "build the app", "start building", or "execute the roadmap".
nextjs-framer-motion-animations
IncludedAdds production-safe Motion for React or Framer Motion animations to Next.js apps, including reveal, hover and tap micro-interactions, whileInView, stagger, AnimatePresence, layout and layoutId transitions, reorder, scroll-linked UI, and lightweight route-content transitions. Use when the user asks to add, refactor, or debug Motion or Framer Motion in App Router or Pages Router codebases, especially around server/client boundaries, reduced motion, LazyMotion, bundle size, hydration, or route transitions. Avoid for GSAP-style timelines, WebGL or 3D scenes, heavy scroll storytelling, or CSS-only effects unless Motion is explicitly requested.