Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.patchline.ai/llms.txt

Use this file to discover all available pages before exploring further.

This page installs the docs MCP — a small, public, read-only server that lets your AI search and read the Patchline documentation. It is not the main Patchline MCP that gives your AI access to your catalog, releases, and tools. For that, see Patchline MCP overview and Install the MCP server. Most users install both.
Connect the docs MCP to Claude, Cursor, Codex, or VS Code with one click — your AI tool gains native search across the entire Patchline knowledge base. No more copy-paste, no more “I can’t read that URL,” no more re-explaining how the product works to your AI.

​
What this installs

The Patchline docs MCP at https://docs.patchline.ai/mcp. Once installed, your AI tool gets two native tools it can call any time:
  • search_patchline — semantic search across the full Patchline knowledge base (returns the relevant docs sections with links).
  • query_docs_filesystem_patchline — read any docs page directly, search with rg, list the tree with tree and ls.
The AI tool decides when to use them. You just ask your normal questions — “how does Patchline pricing work?”, “how do I plan a release?”, “what does the Audience tab show?” — and the answers come grounded in the real docs.
This is the docs MCP (search + read the documentation). It’s different from the product MCP at https://www.patchline.ai/api/mcp/v1, which gives AI tools access to your actual Patchline account (catalog, releases, etc.). See MCP overview for the full distinction.

​
Cursor — one-click install

Install in Cursor → Click that link on the same machine running Cursor — Cursor opens and asks you to confirm the install. Done in 5 seconds. Manual config (if the deeplink doesn’t open): 
{
  "mcpServers": {
    "patchline-docs": {
      "url": "https://docs.patchline.ai/mcp"
    }
  }
}
Paste into Settings → MCP → + Add new MCP server.

​
VS Code — one-click install

Install in VS Code → Manual config: create .vscode/mcp.json in your workspace: 
{
  "servers": {
    "patchline-docs": {
      "type": "http",
      "url": "https://docs.patchline.ai/mcp"
    }
  }
}

​
Claude Code — one-line install

claude mcp add --transport http patchline-docs https://docs.patchline.ai/mcp
Verify with: 
claude mcp list
You should see patchline-docs in the list.

​
Claude Desktop

Open Settings → Developer → Edit Config (claude_desktop_config.json) and add: 
{
  "mcpServers": {
    "patchline-docs": {
      "url": "https://docs.patchline.ai/mcp",
      "transport": "http"
    }
  }
}
Restart Claude Desktop. The Patchline docs become a native connector. Or via Claude’s Connectors UI:
  1. Go to claude.ai/settings/connectors.
  2. Click Add custom connector.
  3. Name: Patchline Docs. URL: https://docs.patchline.ai/mcp.
  4. Click Add.

​
ChatGPT

ChatGPT MCP support is in flight. In the meantime, ChatGPT Plus / Team users can paste any https://docs.patchline.ai/<page>.md URL into a chat with Browse enabled and ChatGPT will read it directly.

​
Test that it worked

After installing, ask your AI any of these:
  • “What can you tell me about Patchline?”
  • “How does the Aria release planner work?”
  • “What’s the Patchline pricing structure?”
  • “Where does the Audience tab pull data from?”
If it answers with citations to docs.patchline.ai/... pages, you’re connected.

​
Why this is better than URL fetching

Most AI tools can’t reliably fetch web URLs. ChatGPT free tier doesn’t browse. Claude needs Web search toggled on per chat. Cursor doesn’t have browse at all. An MCP install solves it permanently — the docs become a native, always-on tool the AI can use without any fetching. You install once. Every chat after that, your AI knows Patchline.

​
What’s NOT in the docs MCP

The docs MCP only searches and reads the documentation. It does NOT:
  • Read your actual Patchline account
  • Add artists or upload tracks
  • Run releases or generate visuals
For those actions, install the product MCP at https://www.patchline.ai/api/mcp/v1 — see MCP overview. Most users install both: the docs MCP to learn Patchline, the product MCP to act on their data.