arxdsilva/agent-office / .claude/skills/parallel-work
parallel-work
VerifiedSplit a multi-area agent-office task across parallel subagents per .claude/rules/PARALLEL_WORK.md — contract first, disjoint file ownership, lead-only shared config, phased integration. Use for any task touching two or more of assets, core, adapters, desktop, renderer logic, renderer adapters or docs, or whenever the user asks to work in parallel.
Parallel work (lead checklist)
- Map areas → agents. List the areas the task touches; one agent per area/layer (hexagon). ≤ 5 running at once.
1b. Pick each agent's model (rule §6): the cheapest competent tier —
haikufor mechanical work,sonnetfor implementation against a written contract,opusfor protocol/lifecycle/security/core-refactor work. Escalate one tier after two failed gates. - Contract first (you, alone):
- Shared types and docs (
docs/*.md,packages/shared,packages/web/src/app/office-types.ts), port interfaces, IPC/HTTP shapes — additive only. pnpm addevery dependency the agents will need.- All shared-config edits (package.json, Makefile, vitest/tsconfig, depcruise, CI, gitignore/gitattributes).
- Shared types and docs (
- Launch producers in parallel (assets, backend, pure logic), then consumers (they may start against contract + fixtures). Background agents; one message with all launches.
- Relay every cross-agent fact via
SendMessage; broadcast contract changes to every affected agent. - Integrate and verify yourself after each phase:
make check,make build test-ui,pnpm arch, screenshots for UI/assets. Route failures to the owner. - Don't enter plan mode while agents run (it pauses them); if you did, resume each with
SendMessage.
Agent prompt template
Model: <haiku|sonnet|opus> — <one-line reason> (set the Agent `model` parameter to match)
You are agent <ID> on the agent-office repo (/Users/asilva/youtube/workspace). Read first (authoritative):
- <plan file §…>, docs/<contract>.md §…, AGENTS.md, .claude/rules/{ARCHITECTURE,TESTING,PRIVACY_SECURITY,SCENE,PARALLEL_WORK}.md
## Your job
<scope, numbered, with acceptance criteria>
## Rules
- Owns ONLY: <files/globs>. Must NOT touch: <files/globs> (owned by <agents>), package.json, lockfile, Makefile, configs.
- No pnpm add/install (report missing deps). No git add/commit/push.
- Strict TDD; keep `pnpm arch` green; run only: <targeted commands>.
- Strip ELECTRON_* env vars when spawning Electron/Blender (VS Code exports ELECTRON_RUN_AS_NODE=1); use make targets.
- If a producer isn't done, develop against the contract + fixtures and report.
## Final report (under N words)
<what to report: APIs, test counts, deviations, open issues>
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 466k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 196k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |