mirror of
https://github.com/andrewyng/openworker.git
synced 2026-08-30 22:53:41 +00:00
Merge 1f2095d030 into 9e145d9ceb
This commit is contained in:
@@ -68,7 +68,7 @@ Unattended runs never self-approve: their asks park in an inbox until a human an
|
||||
|
||||
- **Produce real deliverables** - documents, spreadsheets, reports, and web pages land as files you can open and share.
|
||||
- **Work from Slack** - mention `@OpenWorker` in a channel; a session opens on your desktop, the work happens with your tools, and the answer comes back as a thread reply.
|
||||
- **Use your everyday tools** - 25+ integrations including GitHub, Slack, Jira, Notion, Linear, HubSpot, Outlook, monday.com, Gmail, and Google Calendar, plus your **terminal and local files**. Any tool reachable over [MCP](https://modelcontextprotocol.io/) plugs in too, with per-tool control.
|
||||
- **Use your everyday tools** - 25+ integrations including GitHub, Slack, Jira, Notion, Linear, HubSpot, Outlook, monday.com, Gmail, and Google Calendar, plus your **terminal and local files**. Any tool reachable over [MCP](https://modelcontextprotocol.io/) (see [MCP Setup Guide](docs/mcp-setup.md)) plugs in too, with per-tool control.
|
||||
- **Run on a schedule** - automations for recurring work: a morning brief, a weekly report, a standing watch over a channel. Runs land in the app with full transcripts.
|
||||
- **Ask before acting** - writes, sends, and shell commands are approval-gated, with an optional auto-approve mode that still escalates anything uncertain - see [Governed by design](#governed-by-design).
|
||||
|
||||
|
||||
@@ -0,0 +1,50 @@
|
||||
# Adding MCP Servers to OpenWorker
|
||||
|
||||
OpenWorker supports integrating with Model Context Protocol (MCP) servers via configuration files. This allows you to extend the agent's capabilities with tools from any standard-compliant MCP server.
|
||||
|
||||
## Configuration Files
|
||||
OpenWorker looks for `mcp.json` in two locations:
|
||||
1. **Workspace Specific:** `<workspace>/.coworker/mcp.json` (overrides global settings).
|
||||
2. **Global:** `~/.config/coworker/mcp.json`
|
||||
|
||||
## Configuration Format
|
||||
The configuration follows the standard MCP configuration format:
|
||||
|
||||
```json
|
||||
{
|
||||
"mcpServers": {
|
||||
"server-name": {
|
||||
"command": "node",
|
||||
"args": ["/path/to/server/index.js"],
|
||||
"env": {
|
||||
"API_KEY": "your_key"
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
## Key Features
|
||||
- **Auto-detection:** The system automatically identifies the transport layer (`stdio` or `HTTP` / `SSE`) based on your configuration.
|
||||
- **Variable Expansion:** Use `${VAR}` syntax in `command`, `args`, `env`, `url`, and `headers`. These are resolved at load time using the system's environment and `.env` files.
|
||||
- **Tool Filtering:** Optionally restrict which tools from an MCP server are exposed to the agent:
|
||||
- `include_tools`: Explicit list of allowed tool names.
|
||||
- `exclude_tools`: List of blocked tool names.
|
||||
|
||||
## Quick Start Example
|
||||
To add a local server, create `.coworker/mcp.json` in your project root:
|
||||
```json
|
||||
{
|
||||
"mcpServers": {
|
||||
"my-tool": {
|
||||
"command": "python",
|
||||
"args": ["/path/to/server.py"],
|
||||
"env": {
|
||||
"DEBUG": "true"
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
The agent will recognize these tools immediately on the next run.
|
||||
Reference in New Issue
Block a user