菜单
Implement an already-approved plan from swarm-report/<slug>-plan.md. Routes each task to the executing agent that owns its file scope (frontend / backend / devops / mobile), runs the tests, writes an implementation report. Use AFTER /plan, never before.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/ayusavin/labs-harness --skill build命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Turn a design (Figma file or an exported reference deck) into a build ladder — inventory every frame, extract tokens and atoms, find the gaps the designer still owes you, and write a /build-ready plan plus the .claude/design-gate.json rungs that /visual-verify will check. Use BEFORE writing any UI code from a design. Not for features without a design.
Install this mini dev-loop harness into a target project. Copies the agents, skills (plan/build/review/debug), the test-gate hook, settings, and AGENTS.md into the given project path, then seeds a Memory Bank index if none exists. Run this from a clone of the harness repo — pass the absolute path to the project you want to equip.
Check the built UI against the design, rung by rung — capture the real rendered component (Storybook via Playwright on web, the debug preview gallery on an iOS simulator or Android emulator), diff it against the 1:1 design export, and write the verdict the visual-gate Stop hook reads. Use after /build lands a rung, and whenever the gate blocks. Requires .claude/design-gate.json.
基于 SOC 职业分类
正在显示 SKILL.md
| name | build |
| description | Implement an already-approved plan from swarm-report/<slug>-plan.md. Routes each task to the executing agent that owns its file scope (frontend / backend / devops / mobile), runs the tests, writes an implementation report. Use AFTER /plan, never before. |
ORCHESTRATOR. You route work to executing agents by file scope. No code edits in the main loop.
/build <slug>
swarm-report/<slug>-plan.md. Missing → abort:
"Run /plan \"<feature>\" first."affected_files. Match each against the Executing
table in AGENTS.md (frontend / backend / devops / mobile). Group files by agent.
No match for a file → ask the user which agent owns it.Task call each, in parallel when a feature
spans several layers (single message, N calls, subagent_type: general-purpose).
Prompt each:
Answer TERSE. Read
.claude/agents/<scope>.mdand follow it exactly. Plan: swarm-report/-plan.md. Your scope: . If one agent's output changes a contract another needs (e.g. backendapi_changes), pass that note into the dependent agent's prompt — or run backend first, then frontend.
tests_result. Any status: blocked or failing test →
do NOT claim success. Report the failures verbatim and stop. Offer /build <slug>
retry after the fix.swarm-report/<slug>-build.md: per-agent changed files, test commands +
real results, cross-layer notes.