Granoflow MCP Server
Exposes Granoflow's local REST API and CLI as tools for AI agents, IDEs, and automation, enabling task and project management via natural language.
README
Granoflow MCP Server
MCP server for Granoflow: exposes the Granoflow Local HTTP API as tools for AI agents, IDEs, and automation.
This server is intentionally thin. It does not own Granoflow business logic, database access, app orchestration, or release workflows. It resolves a local API endpoint, forwards structured requests to the running Granoflow app, and returns predictable MCP tool results.
Requirements
- Node.js 20 or newer.
- A running Granoflow app with the Local HTTP API enabled.
The default Granoflow API URL is:
http://127.0.0.1:56789
You can override it with:
export GRANOFLOW_API_BASE_URL="http://127.0.0.1:56789"
export GRANOFLOW_API_TOKEN="..."
The MCP server can keep non-secret local connection defaults in:
~/.config/granoflow-mcp/config.json
Set GRANOFLOW_MCP_CONFIG_PATH to use a different config path for tests,
temporary setups, or advanced local installs. API tokens are not stored in this
file; keep GRANOFLOW_API_TOKEN in the MCP client environment.
Install
npm install -g @granoflow/mcp-server
For a user-facing setup walkthrough, see Granoflow MCP User Install And Demo Guide.
For maintainers, see Granoflow MCP Release Checklist.
Release Branch Policy
developis the active integration branch. It may contain unverified or unreleased changes.mainis the npm release branch. Publish@granoflow/mcp-serverlatest only frommain.- Merge or fast-forward
developintomainonly after release preflight passes.
For local development:
npm install
npm run build
node dist/index.js
Verify an installed package without starting an MCP stdio session:
npx -y @granoflow/mcp-server --version
npx -y @granoflow/mcp-server --help
Before publishing a release, verify the package contents:
npm run release:preflight
Tools
Initial tools:
granoflow_setup_statusgranoflow_setup_detect_local_apigranoflow_setup_write_configgranoflow_setup_open_configgranoflow_setup_open_appgranoflow_healthgranoflow_versiongranoflow_capabilitiesgranoflow_ai_agent_toolsgranoflow_task_listgranoflow_task_exportgranoflow_task_validategranoflow_task_importgranoflow_task_creategranoflow_task_create_structuredgranoflow_task_updategranoflow_task_update_structuredgranoflow_task_completegranoflow_task_finishgranoflow_task_resolvegranoflow_project_listgranoflow_project_resolvegranoflow_project_creategranoflow_project_updategranoflow_project_deletegranoflow_milestone_listgranoflow_milestone_resolvegranoflow_milestone_creategranoflow_milestone_updategranoflow_milestone_deletegranoflow_review_day_showgranoflow_api_request
Prefer the structured task, project, and milestone tools for common resource operations. The JSON payload tools remain available as escape hatches when the running app exposes newer fields before this package has first-class schemas.
Write tools default to dry-run behavior. Ask the tool to write only after you have reviewed the preview or the user has explicitly requested a write. Delete tools also require the current resource title before writing, and refuse linked tasks unless the caller explicitly accepts that impact.
Setup Diagnostics
Use the setup tools when an agent or MCP client needs to connect to a local Granoflow app without hand-editing every setting first:
granoflow_setup_statusreports config path, env/config precedence, token presence, MCP server version, Local HTTP API health, version metadata, capability summary, and local Granoflow process evidence without printing secrets.granoflow_setup_detect_local_apiprobes a small bounded localhost port list only.granoflow_setup_write_configpreviews or writes non-secret config. It defaults to dry-run.granoflow_setup_open_configcreates and optionally opens the config file for manual editing.granoflow_setup_open_apppreviews or opens the installed Granoflow app after user approval. On macOS it tries the formal/Applications/granoflow.apppath before app-name fallbacks. It defaults to dry-run.
When setup status sees a configured localhost API URL that is unreachable, it checks whether a local Granoflow process appears to be running. If not, it returns a warning and asks the agent to confirm before opening the app.
Client Support
This package implements a standard MCP stdio server. The primary compatibility contract is the MCP protocol plus the npm executable:
npx -y @granoflow/mcp-server
Cursor and Codex are the verified client targets for this repository. Other MCP-compatible clients can use the same stdio command shape, but are not part of the routine verification matrix.
Cursor
Add this to .cursor/mcp.json in a project or ~/.cursor/mcp.json globally:
{
"mcpServers": {
"granoflow": {
"command": "npx",
"args": ["-y", "@granoflow/mcp-server"],
"env": {
"GRANOFLOW_API_BASE_URL": "http://127.0.0.1:56789"
}
}
}
}
Codex
Add this to ~/.codex/config.toml:
[mcp_servers.granoflow]
command = "npx"
args = ["-y", "@granoflow/mcp-server"]
[mcp_servers.granoflow.env]
GRANOFLOW_API_BASE_URL = "http://127.0.0.1:56789"
Restart Codex after changing MCP configuration.
Other MCP-Compatible Clients
For clients that support local stdio MCP servers, configure the server with:
{
"type": "stdio",
"command": "npx",
"args": ["-y", "@granoflow/mcp-server"],
"env": {
"GRANOFLOW_API_BASE_URL": "http://127.0.0.1:56789",
},
}
Development
npm install
npm run check
npm run check runs Prettier, ESLint, TypeScript, and Vitest.
Security
- This server does not read or write Granoflow's SQLite/Drift database.
- This server does not run Granoflow app builds, screenshots, release jobs, or scenario orchestration.
- Core operations go through the running app's Local HTTP API.
- API tokens are passed through environment variables and must not be logged.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。