labelhub-platform
AGENTS.md48 starsChanged 15 days agoApache-2.0View on GitHub
AI 编码 agent 的入口(Codex、Claude Code 等)。人类开发者请从 README 开始。 CI(.github/workflows/ci.yml)会执行同样的检查。只要有集成测试被跳过,CI 就判失败。
In this repositoryOne other file it gives its agents.
- design-system-reviewSkill
The file
Files
AGENTS.md
AI 编码 agent 的入口(Codex、Claude Code 等)。人类开发者请从 README 开始。
开始工作前必读
docs/engineering/agent-charter.md:工作方式、Zero-Pause 的激活条件,以及 LH-1~5 项目硬约束。docs/engineering/conventions.md:工程实现约束,包括幂等键格式(§7.1)、Hash 规范化(§7.2)和四个亮点的硬约束(§10)。docs/architecture/labelhub-complete-design-baseline.md:业务事实模型,冲突时以它为准。
不可违反的规则
- append-only 的对象(Schema 版本、提交、Ledger 条目、AI 调用、导出快照)不能更新,也不能删除,对应的 Mapper 也不得提供 update 或 delete 方法。
- AI 的输出只是证据,不是裁决。AI 不能写最终结论。
- 修改 API 时先改
packages/contracts/openapi/labelhub.yaml。前端类型是生成出来的,不要手改。 - 架构层面的变更必须先在
docs/adr/写 ADR,再动代码。 - 无法从文档推断的业务决策,记录到
docs/process/humanpending.md,不要自己猜。
提交前必须全部通过
make verify # 后端:单元测试 + Testcontainers 集成测试
make evals # 离线评测:pytest + 回放复现 smoke 基线(改动 evals/ 时)
pnpm --filter @labelhub/web lint
pnpm --filter @labelhub/web typecheck
pnpm --filter @labelhub/web testCI(.github/workflows/ci.yml)会执行同样的检查。只要有集成测试被跳过,CI 就判失败。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

