DevTask Mcp
Enables AI agents to manage a task board by investigating needs, creating well-specified tasks, and executing them end-to-end.
README
Devtask MCP
Agent-native dev task board — investigate needs into well-specified tasks, execute them end-to-end. Pocock's frontier pattern with spec/slug/dependency fields and scope de-enum.
Combines:
- MCP server (
src/devtask_mcp/) — wraps your kanocifer-chat dev-task API as 6 tools (list_dev_tasks,get_dev_task,get_dev_task_by_slug,create_dev_task,update_dev_task,get_frontier_tasks) - devtask-plan skill (
.claude/skills/devtask-plan/) — investigate a need, interview for spec, create the task - devtask-doit skill (
.claude/skills/devtask-doit/) — execute a task by slug or claim the frontier
安装
1. MCP server
配置到 Claude Code settings(~/.claude.json):
{
"mcpServers": {
"devtask": {
"command": "uv",
"args": [
"run",
"--directory",
"/path/to/DevTaskMcp",
"python",
"-m",
"devtask_mcp.server"
]
}
}
}
或 .mcp.json 项目根:
{
"devtask": {
"command": "uv",
"args": [
"run",
"--directory",
"/path/to/DevTaskMcp",
"python",
"-m",
"devtask_mcp.server"
]
}
}
2. 环境变量
cp .env.example .env
# 编辑 .env 填 DEVTASK_API_KEY + DEVTASK_API_BASE
3. Skills
技能文件在 .claude/skills/ 下,项目内自动加载。若想全局使用:
# 把软链或复制到全局 skills 目录
ln -s /path/to/DevTaskMcp/.claude/skills/devtask-plan ~/.claude/skills/devtask-plan
ln -s /path/to/DevTaskMcp/.claude/skills/devtask-doit ~/.claude/skills/devtask-doit
使用
/devtask-plan # 调研发明一个任务
/devtask-doit # 执行 frontier 里下一个
/devtask-doit task-42 # 执行指定 slug 的任务
任务模型
| 字段 | 语义 |
|---|---|
slug |
task-N 人类可读 ID,自增单调 |
acceptance_criteria |
验收条件,doit 时逐条自检 |
constraints |
红线(不能动哪些文件) |
context_pointers |
相关代码路径,省去文件探索 |
for_agent |
是否 agent 可执行 |
blocked_by |
依赖的任务 slug 列表 |
scope |
<层>-<技术> 自由格式 |
结构
DevTaskMcp/
├── .claude-plugin/marketplace.json # 插件市场清单
├── skills/
│ ├── devtask-plan/SKILL.md # 技能:调研 → 生成 spec → 创建任务
│ └── devtask-doit/SKILL.md # 技能:按 slug 执行任务
├── src/devtask_mcp/ # MCP server Python 包
│ ├── __init__.py
│ ├── client.py # HTTP 封装,拆 envelope
│ ├── models.py # Pydantic 模型 + 中文 enum
│ └── server.py # FastMCP 6 tool 注册
├── pyproject.toml
└── README.md
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.