architecture-dev-handoff
Generate Component Development Handoff documents from architecture documentation. Creates per-component, audience-segmented (Dev/QA/Ops) implementation guides in an 8-section condensed format with deliverable assets (OpenAPI specs, DDL scripts, Kubernetes manifests). Invoke when handing off a component for implementation, when a dev team needs implementation specs, or when generating development handoff documents.
What this skill does
# Architecture Dev Handoff Skill — Orchestrator
## When This Skill is Invoked
Manually activated when users request component handoff document generation. It is NOT automatically triggered.
**Activation Triggers:**
- "Generate handoff for [component name]"
- "Create development handoff"
- "Hand off [component] to development"
- "Generate implementation spec for [component]"
- "Create dev handoff for all components"
- "Generate handoff documents"
- "Create handoff docs"
- "Prepare components for development"
- `/skill architecture-dev-handoff`
- `/skill architecture-dev-handoff all`
- `/skill architecture-dev-handoff [component-name]`
**NOT Activated For:**
- General architecture documentation tasks (use `architecture-docs` skill)
- Compliance generation (use `architecture-compliance` skill)
- Component index maintenance (use `architecture-component-guardian` skill)
- Peer review (use `architecture-peer-review` skill)
---
## Architecture — Orchestrator + Sub-Agent
As of v3.7.0, this skill is an **orchestrator**. Per-component generation runs in isolated sub-agent contexts via `agents/generators/handoff-generator.md` (model: sonnet) and `agents/builders/handoff-context-builder.md` (model: sonnet, added in v3.13.0). The orchestrator stays in main context; sub-agents do the heavy per-component reads.
- **Main context budget**: ~80–120 KB per invocation, flat regardless of how many components are selected.
- **Sub-agent budget**: ~25–40 KB per spawn (per-component payload + handoff template + section-extraction guide + ranged slice of asset guide).
- **Parallelism**: sub-agents spawn in batches of N (default 4, configurable via `--parallelism N`, capped at 8), one message per batch. Default raised from 2 in v3.13.0 — sub-agents are read-only on independent payloads, so 4-way fan-out cuts wall-clock by ~50% on 8-component runs without extra rate-limit risk on standard tiers.
Why this matters: previous versions re-read 7 shared architecture files plus the full `adr/` directory once per component. For a 10-component run that was ~750–2,750 KB of redundant reads. The orchestrator now reads them **exactly once** and slices per component into payload files that sub-agents consume.
See `PAYLOAD_SCHEMA.md` for the orchestrator → sub-agent contract.
---
## Purpose
This skill generates **Component Development Handoff** documents — one per component — that give development teams everything needed to implement a component without reading the full ARCHITECTURE.md suite.
**Scope:** C4 Level 2 (Container) components only. C4 Level 1 (System) descriptors are excluded — they describe system boundaries, not implementable units.
**Output location:** `handoffs/` directory at the project root
**Per-component output** (all names use **lowercase kebab-case** — no spaces, no uppercase, no underscores):
- `handoffs/NN-<component-name>-handoff.md` — 8-section audience-segmented handoff document (Dev / QA / Ops tracks; TEMPLATE_VERSION 2.0.0)
- `handoffs/assets/NN-<component-name>/` — scaffolded deliverable assets (OpenAPI spec, DDL, Kubernetes manifests, etc.)
- `handoffs/README.md` — managed index updated after each generation
The `<component-name>` slug is derived from the component file name in `docs/components/` (e.g., `docs/components/03-payment-service.md` → `handoffs/03-payment-service-handoff.md`).
---
## Template Format Preservation Policy
**CRITICAL REQUIREMENT**: Generated handoff documents MUST preserve template format exactly.
**Strict Preservation Rules:**
1. ✅ **ONLY replace explicit `[PLACEHOLDER]` tokens** — text inside `[...]` brackets
2. ✅ **PRESERVE ALL other text exactly** — including formatting, headers, structure
3. ❌ **NEVER transform template content** — no custom prose restructuring, no reformatting
4. ✅ **GAP DETECTION** — if data is not found in architecture docs, write `[NOT DOCUMENTED]` and add entry to section D1 (Open Questions & Assumptions)
5. ❌ **NEVER invent** values not stated in the architecture documentation
These rules apply to both the orchestrator (payload construction) and the sub-agent (template filling).
---
## Strict Source Traceability Policy
All handoff documents must trace data back to architecture docs.
1. ✅ **ONLY extract data** that explicitly exists in the architecture docs
2. ✅ **CITE source files** (e.g., `docs/components/01-api-gateway.md`, `docs/07-security-architecture.md`)
3. ✅ **NEVER infer or guess** values not stated in the docs
4. ❌ **NO INFERENCE** — even if industry standards suggest a value
5. ✅ When data is missing: write `[NOT DOCUMENTED — add to docs/NN-file.md]` and record in section D1
---
## Documentation Fidelity Policy
**CRITICAL REQUIREMENT**: Every generated handoff document and asset file MUST be an exact representation of what is specified in the architecture documentation. No more, no less.
**Rules:**
1. ✅ **EXACT MATCH** — every value in a generated handoff or asset (endpoint, field, schema, resource limit, env var, port, path, version, etc.) must come verbatim from the architecture docs
2. ❌ **NO DEFAULTS** — do not fill in default/fallback values in either handoff documents or asset files. If the value is not documented, use `[NOT DOCUMENTED — add to <source-file>]` in handoff sections or `# TODO: [NOT DOCUMENTED]` in asset files
3. ❌ **NO EXTRA FIELDS** — do not add fields, endpoints, schemas, columns, or resources beyond what is explicitly documented
4. ❌ **NO OMISSIONS** — every documented value relevant to the section or asset type MUST appear in the generated output
5. ✅ **COMPLETENESS CHECK** — after generating each handoff section and asset, verify it against the source documentation: every documented item has a corresponding entry, and every entry traces back to a documented item
---
## Orchestration Workflow
### Step 0 — Resolve Plugin Directory
Before any workflow, resolve the absolute path to the plugin installation so sub-agents can read the handoff template, section-extraction guide, asset-generation guide, and asset index.
**Step 0a — Glob (dev/local mode)**:
Glob for: `**/{sa-skills,solutions-architect-skills}/skills/architecture-dev-handoff/SKILL.md`
The brace expansion matches both marketplace installs (`sa-skills/` in `~/.claude/plugins/cache/...`) and local dev clones (historical repo folder `solutions-architect-skills/`). If found, strip `/skills/architecture-dev-handoff/SKILL.md` from the result to get `plugin_dir`.
**Step 0b — Marketplace fallback**:
If Glob returns nothing, set:
```
plugin_dir = ~/.claude/plugins/marketplaces/shadowx4fox-solution-architect-marketplace
```
**Step 0c — (removed in v3.13.0)**
Earlier versions mirrored four reference files (`HANDOFF_TEMPLATE.md`, `SECTION_EXTRACTION_GUIDE.md`, `ASSET_GENERATION_GUIDE.md`, `assets/_index.md`) to `/tmp/handoff-plugin-refs/` so sub-agents could Read them despite permission scoping. As of v3.13.0 those references are bundled directly into `agents/generators/handoff-generator.md` (sub-agent system prompt). Sub-agents no longer Read any plugin file at runtime, so the staging fallback is dead overhead and has been removed.
`plugin_dir` is still resolved in Steps 0a/0b — the orchestrator uses it to invoke `bun [plugin_dir]/skills/architecture-dev-handoff/utils/{prepare-payload-dir,manifest}.ts`. If a bun helper invocation fails with a permission error, check that `Bash(bun *)` is granted in project settings.
### Phase 1: Initialization
**Step 1.0: Parse orchestrator flags**
Inspect the user's invocation for flags. Default values apply when a flag is absent.
| Flag | Default | Validation | Purpose |
|------|---------|------------|---------|
| `--parallelism N` | `4` | integer 1–8; on out-of-range value, warn (`⚠ --parallelism N out of range; clamping to <1\|8>`) and clamp | Stage 5A batch size (handoff-document fan-out) |
| `--asset-parallelism N` | `4` | integer 1–8; on out-of-range value, warn (`⚠ --asset-parallelism N out of range; clamping to <1\|8>`) and clamp | Stage 5BRelated 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.