pipeflow-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@pipeflow-mcpcreate a pipeline for bug fix with stages: reproduce, diagnose, fix"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
pipeflow-mcp
PipeFlow MCP Server —— AI Agent 任务的被动状态跟踪容器。
注意:本项目是实验 / demo 性质的学习作品,作者发布它用于代码评审与学习,真实场景下的实用价值尚未确证,不保证有用。
这是什么
pipeflow-mcp 是一个基于 Model Context Protocol (stdio JSON-RPC) 的 MCP 服务器。它把 agent 的工作流执行过程记录到一个由 FSM 驱动的 pipeline 账本里,提供:
FSM 状态机:5 状态 × 5 action,
pending / running / succeeded / failed / skipped,由transitions库实现审计完整性:
reset保留所有 annotations + reports,delete是 soft-delete,events表 append-only重试与时间预算:retry 计数器 lifetime,stage-level
time_budget_seconds派生time_status16 个
pipeflow_*工具:声明 / 状态推进 / 读取 / 注解 / 报告 / 仪表盘 / 健康检查,其中pipeflow_stage_advance是唯一的状态变更入口Cursor 分页:4 个 list 工具共享
base64-urlsafe(last_id + filter_fingerprint)不透明 cursor,filter 改变即 cursor 失效静态 HTML 仪表盘:导出产物含 7 个 section —— Pipeline Overview / Decision Chain / Reflection Trace / Iteration Compare / Metric Analysis / Deliverables / Events Stream,Jinja2 autoescape 防 XSS,导出路径白名单
错误脱敏:客户端消息只含
pipeline_id/stage_id/ 错误类名,SQL 片段 / 文件路径 / 堆栈仅走 stderr 结构化日志Schema 校验:5 个 ReportType ——
decision / reflection / metric_report / deliverable / iteration_summary,对应 JSON Schema,启动时注册类型化响应:8 个 Pydantic v2 view model,FastMCP
outputSchema自动填充字段契约,extra='forbid'锁定公开 JSON 契约
完整 API 文档见 docs/api.md。
Related MCP server: Subconductor
状态
指标 | 值 |
版本 | v0.1.0 |
MCP 工具 | 16 个 |
测试 | 584 passing,unit + integration + e2e + stdio smoke |
静态检查 |
|
Python | 3.11 - 3.14,运行时最低 3.11,CI 矩阵 3.13 + 3.14 |
存储 | SQLite WAL,单写者假设,多进程并发写不在设计范围内 |
安装与运行
MCP 客户端
从 GitHub 直接安装:
uvx --from git+https://github.com/AstralVoidZ/pipeflow-mcp pipeflow-mcp或者从本地源码 checkout 启动:
git clone https://github.com/AstralVoidZ/pipeflow-mcp.git
cd pipeflow-mcp
uv venv --python 3.14 .venv
uv pip install -e ".[dev]"
uvx --from . pipeflow-mcp # dev 模式Server 通过 stdio 监听 JSON-RPC,stdout 保留给 JSON-RPC,所有日志走 stderr。
客户端配置示例
Claude Code,~/.claude/mcp_servers.json:
{
"mcpServers": {
"pipeflow": {
"command": "uvx",
"args": ["--from", "git+https://github.com/AstralVoidZ/pipeflow-mcp", "pipeflow-mcp"],
"env": {
"PIPEFLOW_DB_PATH": "~/.pipeflow_mcp/state.db",
"PIPEFLOW_LOG_LEVEL": "INFO"
}
}
}
}opencode,项目级 .opencode/config.json:
{
"mcp": {
"pipeflow": {
"type": "local",
"command": ["uvx", "--from", "git+https://github.com/AstralVoidZ/pipeflow-mcp", "pipeflow-mcp"],
"environment": {
"PIPEFLOW_DB_PATH": "~/.pipeflow_mcp/state.db",
"PIPEFLOW_LOG_LEVEL": "INFO"
},
"enabled": true
}
}
}工具矩阵
Category | Tool | 说明 |
Declare |
| 一次性创建 pipeline + stages,支持 |
| 追加 stage 到已存在 pipeline | |
State |
| 唯一状态变更入口,start / complete / fail / skip / retry |
Read |
| 单 pipeline 当前状态 |
| 完整快照,含 reports 聚合 + time_status | |
| pipelines 列表,metadata_filter + cursor | |
| events 流,stage_id / event_types / since / until + cursor | |
Annotation |
| upsert 注解,决策记录 |
| 注解索引,不含 data,3 层梯度 | |
| 注解完整 payload | |
Report |
| 提交 5 种类型报告,schema 校验 |
| 报告列表,type / iteration / cursor | |
Dashboard |
| append-only reset,保留注解 + 报告 |
| soft-delete,list 排除,history 保留 | |
| 7 段 HTML 仪表盘导出,白名单路径 | |
Health |
| liveness + readiness + uptime, |
完整参数 / 返回 schema / 错误清单见 docs/api.md。
配置
pipeflow-mcp不接受任何 CLI 参数,无--db-path之类的 flag。所有配置通过环境变量传入,客户端在 MCP server 配置的env块里设置即可。误传--db-path只会在 cwd 留下一个同名空文件。
变量 | 默认值 | 说明 |
|
| SQLite 数据库文件路径,首次启动自动 |
|
|
|
|
| 日志级别, |
|
| stderr 日志格式,设为 |
所有日志写入 stderr,stdout 保留给 JSON-RPC 通信。
架构
5 层严格单向依赖:
protocol (MCP tools / server.py)
→ view layer (Pydantic v2 view models, extra='forbid')
→ persistence orchestration (orchestration/ — advance, create, append, reset, delete, snapshot, annotate, reports, queries, ...)
→ { FSM layer (fsm/machine.py) , Spec layer (yaml_parser, report_registry, result_validator) }
→ data contract layer (models/*.py — dataclasses only)
→ error layer (errors/*.py)侧模块,storage/ (DB + DAO + migrations) · pagination/ (cursor codec) · export/ (Jinja2 + 8 模板) · tools/ (16 薄封装) · config.py (env + logging)。
不变量
stdout = JSON-RPC,stderr = logs,任何 stdout 写入会破坏 MCP 客户端
DAO 是 SQL 边界,orchestration / fsm / spec / tools / pagination / export 永不 import
pipeflow_mcp.storage.dao单写者假设,SQLite WAL 在单 host + 单 writer 下安全,多进程并发写超出设计范围
user_version降级拒绝,永不静默 forward-migrate,启动时抛RuntimeError是正确行为Annotation 列表不含
data,list 返回AnnotationSummary,data只能通过get取得resetappend-only,保留 annotations + reports + 历史 events,只追加type='reset'eventView 单向投影,
tools → views → models是唯一允许方向,反向 import 被test_views_have_no_dao_imports强制禁止
开发
本地验证
ruff check src/pipeflow_mcp/ tests/
mypy src/pipeflow_mcp/
pytest tests/ -v --cov=pipeflow_mcp --cov-fail-under=85打包
本地构建 wheel + sdist 到 release/ 目录,.gitignore 排除。
./scripts/build.sh # clean + lint + type + test + build
./scripts/build.sh --wheel # 只构建 wheel
./scripts/build.sh --check # 跳过 clean,直接基于当前 venv 构建产物 release/pipeflow_mcp-<version>-*.whl 和 release/pipeflow_mcp-<version>.tar.gz,用于个人 GitHub Release 上传。
许可证
MIT,见 LICENSE。
This server cannot be deployed
Maintenance
Related MCP Connectors
Durable agent-to-agent handoffs and shared scratchpad for multi-agent workflows.
Checked language for AI-written agent workflows: limits, cost and data flows known before running.
- mcp-serverOAuthcom.make
Give your AI agents the tools to build, manage, and run automation workflows.
Agentic rails for complex workflows with receipts, fees, and MCP tool access.
Related MCP Servers
- FlicenseNot gradedqualityAmaintenanceEnables AI coding agents to execute formal, stateful workflows with typed contracts, postcondition enforcement, and structured retry logic.1-
- AlicenseAqualityCmaintenanceA persistent state machine and notification system for AI agents to manage complex, multi-step workflows via the Model Context Protocol, preventing context drift by maintaining structured checklists and sending desktop alerts.148 npm2Apache 2.0
- AlicenseNot gradedqualityDmaintenanceOrchestrates AI agents through structured markdown documents, enabling multi-agent workflows with automatic context injection and workflow management.8 npm5MIT
- AlicenseAqualityBmaintenanceEnables deterministic multi-agent pipelines in VS Code and GitHub Copilot, with 25 specialized AI agents and a 9-stage state machine that prevents hallucinated routing.11MIT