devflow
CLAUDE.md4 starsChanged 5 months agoMITView on GitHub
本文件为 AI Agent 开发 DevFlow 自身时提供指导。 DevFlow —— 双 Pipeline 多 Agent 协作工作流 Skill。纯 Markdown + YAML，无编译、零依赖。 SKILL.md 三级渐进加载： - L1（~340 tok）frontmatter —— 始终加载 - L2（~2,400 tok）正文 —— 触发后加载，含 Iron Rules + Step -1→7 + 记忆口诀 - L3 resources/ —— 按需加载，15 个文件分 3 子目录 + 2 根文件（~17,000 tok 合计）
In this repositoryOne other file it gives its agents.
- devflowSkill
The file
Files
CLAUDE.md
本文件为 AI Agent 开发 DevFlow 自身时提供指导。
项目
DevFlow —— 双 Pipeline 多 Agent 协作工作流 Skill。纯 Markdown + YAML，无编译、零依赖。
校验
node tests/devflow-check.js架构
SKILL.md 三级渐进加载：
- L1（~340 tok）frontmatter —— 始终加载
- L2（~2,400 tok）正文 —— 触发后加载，含 Iron Rules + Step -1→7 + 记忆口诀
- L3
resources/—— 按需加载，15 个文件分 3 子目录 + 2 根文件（~17,000 tok 合计）
关键文件
| 文件 | 用途 |
SKILL.md | Skill 入口，6 Iron Rules + Edge Cases + Step -1→7 |
resources/dispatch/rules.md | 调度规则索引（按需加载 dispatch-core/interaction/research/lifecycle） |
resources/dispatch/core.md | 核心调度：模式、派发、测试、DoD、回退、口诀 |
resources/dispatch/interaction.md | PM-Leader 交互纪律 |
resources/dispatch/research.md | 上下文调研规则 |
resources/dispatch/lifecycle.md | 流程管理：版本、文档、Spec、流程闭环、复盘 |
resources/platform-adapters.md | 各平台 Agent 调用语法映射 |
resources/research/schema.md | Research Agent 定义：3 模式 + 2 状态 + Schema |
resources/prompt/guide.md | 提示词工程索引（编译管道 + 路由规则） |
resources/prompt/templates.md | Agent 提示词模板 |
resources/prompt/compression.md | 注入量控制 + 三级压缩 + 故障应对 |
resources/research/taxonomy.md | 知识分类体系 + 映射表 + 未命中处理 |
resources/research/taxonomy-extended.md | 扩展映射表（未命中记录 + 候选） |
resources/research/taxonomy-reorganization.md | 知识领域整理机制（晋升/合并/归档） |
resources/research/search-strategy.md | 搜索操作规范：关键词构造、来源判定、Fetch 要求 |
resources/workflow-config-template.md | 用户项目集成模板 |
README.md + README_en.md | 安装指南、核心规则速查 |
docs/methodology.md | 方法论演进（v0→v3）+ unblind 实战案例 |
docs/contextual-research-plan.md | 调研增强设计文档（参考，非 L3 操作文件） |
docs/devflow-validation.md | 完整性检验文档 |
tests/devflow-check.js | 结构完整性校验脚本 |
by_claude_code_dev_md/progress.md | DevFlow 自身开发进度记录。新 Plan 经同意后 or 实战有新进展时必须更新 |
编辑指南
- 改核心规则 →
SKILL.mdL2，保持 ~700 tok 以内 - 改细节 →
resources/dispatch/rules.md - 新增/改 Agent 类型 →
resources/research/schema.md+platform-adapters.md - 改提示词模板/压缩策略 →
resources/prompt/guide.md - 改集成模板 →
resources/workflow-config-template.md - 分类体系/映射表 →
resources/research/taxonomy.md - 任何结构性变更后跑
node tests/devflow-check.js - 按 semver 更新
SKILL.mdmetadata.version
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.