AI Skill Hub 推荐使用:开源AI工作流:Spec-driven agent harness 是一款优质的Agent工作流。AI 综合评分 7.5 分,在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案,这是一个值得深入了解的选择。
基于Spec驱动的代理器集成, Claude Code和Codex五个代理器管道,提高AI编程效率和准确性。
开源AI工作流:Spec-driven agent harness 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
基于Spec驱动的代理器集成, Claude Code和Codex五个代理器管道,提高AI编程效率和准确性。
开源AI工作流:Spec-driven agent harness 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:npm 全局安装 npm install -g anatomia # 方式二:npx 直接运行(无需安装) npx anatomia --help # 方式三:项目依赖安装 npm install anatomia # 方式四:从源码运行 git clone https://github.com/anatomia-dev/anatomia cd anatomia npm install npm start
# 命令行使用
anatomia --help
# 基本用法
anatomia [options] <input>
# Node.js 代码中使用
const anatomia = require('anatomia');
const result = await anatomia.run(options);
console.log(result);
# anatomia 配置说明 # 查看配置选项 anatomia --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export ANATOMIA_CONFIG="/path/to/config.yml"
Anatomia is a CLI and agent harness for Claude Code and Codex. It scans your codebase — detecting your stack, conventions, and patterns — then runs every change through a five-agent pipeline that saves every artifact: scope, spec, contract, build report, and independent verification. Other harnesses are prompt libraries. This one has an engine.
The cost of building is near zero. The cost of building the wrong thing is the same as it ever was. AI agents rush to implementation — they don't push back on your approach, surface tradeoffs you missed, or record why one design was chosen over another. Anatomia enforces the thinking that fast building skips.
Install globally to use the ana command directly:
npm install -g anatomia-cli
Requires Node.js 22+. To update: npm update -g anatomia-cli
rm -rf .ana # all Anatomia data
rm .claude/agents/ana*.md # Claude Code pipeline agents, if installed
rm -rf .codex/agents # Codex agent manifests, if installed
rm AGENTS.md # generated project summary
Skill directories created by init live under .ana/skills/: coding-standards, testing-standards, git-workflow, deployment, troubleshooting, and conditionally ai-patterns, api-patterns, data-access. Remove the ones init created; keep any you added yourself.
If Anatomia created your CLAUDE.md, remove that too (git blame CLAUDE.md to check). If you ran ana init commit, revert: git revert <commit-hash>. Your source code is never modified.
npm uninstall -g anatomia-cli # remove the CLI
ana init # generate context + agents
ana init commit # persist to git (so teammates get it too)
ana doctor # verify installation is healthy
ana run setup # enrich with your team's knowledge (optional, recommended, ~10 min)
ana run # start working — "hey Ana"
Tell Ana what you want to build. It'll investigate the codebase, surface tradeoffs, and push back if the approach has problems. When the scope is right, it hands off to Plan, Build, and Verify.
initruns scan automatically and works standalone. The pipeline and setup run through Claude Code or Codex, and scan output works with any markdown-aware AI tool.init commitcommits to the artifact branch —staging,develop, or your pre-production branch if one exists, otherwisemain. Check withana config get artifactBranchbefore your first commit. After updating the CLI, runana initagain to refresh scan data and skill detection. Your rules, context, and proof chain are preserved.
| Stage | Command | Role | Produces |
|---|---|---|---|
| Think | ana run | Thinking partner — scope, investigate, advise, push back | scope.md |
| Plan | ana run plan | Architect — design + sealed contract | spec.md + contract.yaml + plan.md |
| Build | ana run build | Builder — implement spec, prove it works | Code + tests + build_report.md |
| Verify | ana run verify | Fault-finder — reads spec and code, skips Build's report | verify_report.md |
| Learn | ana run learn | Proof analyst — runs between cycles | Stronger skills and system improvements |
该项目提供了一个开源的AI工作流,集成了多个代理器,提高了编程效率和准确性,值得关注。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
总体来看,开源AI工作流:Spec-driven agent harness 是一款质量良好的Agent工作流,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | anatomia |
| 原始描述 | 开源AI工作流:Spec-driven agent harness for Claude Code and Codex — five-agent pipeline (scope。⭐29 · TypeScript |
| Topics | workflowagentic-codingaiai-agentsai-codingai-coding-assistanttypescript |
| GitHub | https://github.com/anatomia-dev/anatomia |
| License | MIT |
| 语言 | TypeScript |
收录时间:2026-06-06 · 更新时间:2026-06-06 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端