AI Skill Hub 推荐使用:AI工作流 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。
AI工作流 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。
AI工作流 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 pi-project-workflows # 方式二:npx 直接运行(无需安装) npx pi-project-workflows --help # 方式三:项目依赖安装 npm install pi-project-workflows # 方式四:从源码运行 git clone https://github.com/davidorex/pi-project-workflows cd pi-project-workflows npm install npm start
# 命令行使用
pi-project-workflows --help
# 基本用法
pi-project-workflows [options] <input>
# Node.js 代码中使用
const pi_project_workflows = require('pi-project-workflows');
const result = await pi_project_workflows.run(options);
console.log(result);
# pi-project-workflows 配置说明 # 查看配置选项 pi-project-workflows --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 PI_PROJECT_WORKFLOWS_CONFIG="/path/to/config.yml"
npm install
RUN_INTEGRATION=1 npm test -w packages/pi-workflows
pi install npm:@davidorex/pi-project-workflows
pi install npm:@davidorex/pi-context pi install npm:@davidorex/pi-workflows
npm run build
npm run clean
```bash
Three Pi extensions plus a shared agent-runtime library: typed, multi-step workflow execution via .workflow.yaml specs; schema-driven project state in the substrate directory; behavior monitors that classify agent activity and steer corrections; and a library package that owns everything between "I have a spec" and "I have a typed result."
Schemas are the contract layer. In pi-context, you define what your project tracks by writing JSON Schemas — the tools, validation, and derived state adapt automatically. In pi-workflows, agent steps declare output schemas that enforce the shape of data flowing through the pipeline. In pi-behavior-monitors, JSON pattern libraries define what to detect and how to respond. In pi-jit-agents, agent specs are compiled to typed results with phantom-tool structured output enforcement. The three extensions form a typed loop: project state → workflow input → agent output → validated project state → monitor classification → steering.
| Package | npm | Description |
|---|---|---|
| [@davidorex/pi-context](packages/pi-context/) | npm:@davidorex/pi-context | Schema-driven project state — typed JSON blocks, write-time validation, generic CRUD tools, dynamically derived state. Add a schema, get a new block type with tooling. No code changes. |
| [@davidorex/pi-jit-agents](packages/pi-jit-agents/) | npm:@davidorex/pi-jit-agents | Agent spec compilation and in-process dispatch runtime. Library package (not a Pi extension) that owns loading, compilation, and execution of .agent.yaml specs with phantom-tool structured output enforcement. Consumed by pi-workflows and pi-behavior-monitors. |
| [@davidorex/pi-workflows](packages/pi-workflows/) | npm:@davidorex/pi-workflows | Schema-driven workflow orchestration — YAML specs, DAG execution, typed step types, typed data flow between agents, expression engine, checkpoint/resume. Output schemas are the enforcement boundary between steps. |
| [@davidorex/pi-behavior-monitors](packages/pi-behavior-monitors/) | npm:@davidorex/pi-behavior-monitors | Behavior monitors — autonomous watchdogs that classify agent activity against JSON pattern libraries, steer corrections, and write structured findings. |
Tools: workflow, workflow-list, workflow-agents, workflow-validate, workflow-status, workflow-init
Commands: - /workflow init — scaffold .workflows/ directory - /workflow list — discover and select a workflow to run - /workflow run <name> — execute a workflow (tab-completes with discovered workflow names) - /workflow resume <name> — resume from checkpoint - /workflow validate [name] — validate workflow specs - /workflow status — show workflow vocabulary and discovery - /workflow help — show available subcommands
Keybindings: Ctrl+H pause, Ctrl+J resume
Key concept: Workflows are .workflow.yaml specs with typed data flow between steps. Each step runs as a subprocess with its own context window. The DAG planner infers parallelism from ${{ steps.X }} expression references and context declarations. Agent steps support context: [stepName] to inline prior step narrative text into the dispatch prompt, complementing expression-based structured data flow. The monitor step type integrates behavior classification as a verification gate. Bundled agents, schemas, and templates ship with the package; users override by placing files in .pi/agents/, .pi/templates/.
npm test -w packages/pi-context npm test -w packages/pi-workflows npm test -w packages/pi-behavior-monitors
npm run lint # check for lint issues npm run format # auto-fix formatting npm run check # lint + typecheck
高质量的AI工作流项目,值得关注
该工具未明确声明开源协议,商业使用前请联系原作者确认授权范围,避免侵权风险。
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。
总体来看,AI工作流 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 | pi-project-workflows |
| 原始描述 | 开源AI工作流:A set of pi extensions that allow users and llm's to craft customized project an。⭐19 · TypeScript |
| Topics | AI工作流TypeScript |
| GitHub | https://github.com/davidorex/pi-project-workflows |
| 语言 | TypeScript |
收录时间:2026-05-25 · 更新时间:2026-05-30 · License:未公布 · 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