Cursor reads MCP server definitions from a JSON file that you can place either at the project level (version-controlled, shared with your team) or in your global Cursor settings (shared across all projects on your machine). Either way, the config shape is the same: anDocumentation Index
Fetch the complete documentation index at: https://mintlify.com/bruhsb/paperclip-mcp/llms.txt
Use this file to discover all available pages before exploring further.
mcpServers object containing a paperclip entry with command, args, and env keys. Once saved and reloaded, Cursor exposes all 104 Paperclip tools to its AI chat panel.
Config file location
| Scope | Path |
|---|---|
| Project | .cursor/mcp.json in your project root |
| Global | Cursor User Settings → MCP (GUI), or ~/.cursor/mcp.json |
Environment variables
| Variable | Required | Description |
|---|---|---|
PAPERCLIP_API_KEY | Yes | Bearer token for API authentication |
PAPERCLIP_API_URL | Yes | Base URL of the Paperclip API, e.g. http://127.0.0.1:3100 |
PAPERCLIP_AGENT_ID | Yes | UUID of the agent running this MCP server |
PAPERCLIP_COMPANY_ID | Yes | UUID of the company (used for company-scoped endpoints) |
PAPERCLIP_RUN_ID | No | Heartbeat run ID — injected automatically by Paperclip during agent runs |
PAPERCLIP_REQUEST_TIMEOUT_MS | No | HTTP request timeout in milliseconds (default: 30000) |
Configuration
Open (or create) the target config file and paste the block for your preferred variant.- npm (recommended)
- Docker / Podman
Requires Node.js ≥ 22 on the host. No container runtime needed.
Reload and verify
Reload MCP servers in Cursor
Open the Command Palette (
Ctrl+Shift+P / Cmd+Shift+P) and run Cursor: Reload MCP Servers. Alternatively, restart Cursor entirely.For Paperclip-orchestrated heartbeat runs,
PAPERCLIP_RUN_ID is injected automatically into the process environment — you do not need to add it to the config file.See also
- Troubleshooting — server not connecting, auth errors, network issues
- Auth keys — how to obtain and rotate your
PAPERCLIP_API_KEY - Docker / Podman guide — building and running the container image locally