Patchline MCP — your music data in your AI

Your entire catalog, routed directly into your AI. The Patchline MCP server connects Cursor, Codex, Claude Desktop, Claude Code, VS Code — any MCP-compatible client — straight to your Patchline account. Catalog, artist intelligence, releases, playlists, smart links, music data: every Aria capability becomes a callable tool your AI can use without leaving its window. Endpoint: https://www.patchline.ai/api/mcp/v1 Auth: OAuth 2.0 with PKCE — your credentials never touch the AI client. Install: one command or one config snippet. Free to install on every tier.

Why this matters

Stop copy-pasting ISRC codes. Stop screenshotting your Spotify dashboard into ChatGPT. Stop re-explaining your catalog every chat. Patchline routes your music business into the AI tool you already use:

In Cursor / Codex / VS Code: “Find playlists for my new track” — the AI calls Patchline directly, pitches come back ranked, with reasoning.
In Claude Desktop / Claude Code: “What’s Mira’s audience trajectory and which release should I prioritize?” — Aria’s full knowledge of your roster surfaces in the chat.
In any MCP-compatible client: the same tools, same data, same OAuth.

This is not a chatbot you switch to. It’s a set of tools your existing AI gains.

What’s actually exposed

Grouped by category. The MCP server exposes the live product surface — the exact list and schemas refresh with the product. Connect a client and run tools/list for the authoritative inventory.

Dispatch — always call this first

Catalog

Artists & roster

Releases & projects

Playlists

Smart links

Music data

AI generation

Additional tool families

The MCP server also exposes additional tools that evolve with the product:

Projects & campaigns — create release projects, build campaign plans, fetch artist context.
Public surface links — generate drop links, pitch links, media kits, briefings, and shareable asset links.
Catalog upload — get signed upload URLs and confirm finished uploads from agents that fetch audio.
Similarity — find tracks with similar audio characteristics to a given catalog asset.

For the canonical, current list with full schemas, connect an MCP client and run tools/list.

Install

Install the MCP server → Cursor, Codex, Claude Desktop, Claude Code, VS Code — each has a one-command or one-snippet install. Cursor and VS Code support a 1-click install button. The fastest path:

{
  "mcpServers": {
    "aria": {
      "url": "https://www.patchline.ai/api/mcp/v1"
    }
  }
}

Paste that into Cursor → Settings → MCP → + Add new MCP server. First tool call triggers OAuth.

How auth works

Patchline’s MCP server uses OAuth 2.0 with PKCE — the same flow Spotify, Google, GitHub use:

You add the Patchline MCP URL to your client’s config.
First tool call opens a browser to patchline.ai.
You sign in (or are already signed in) to Patchline.
You approve the AI client’s access.
The client receives a bearer token and uses it for every call.

Your credentials never touch the AI client. Tokens are bound to the client, revocable from Patchline Settings, and expire on a schedule. The MCP server publishes OAuth metadata at:

https://www.patchline.ai/.well-known/oauth-protected-resource

Per RFC 9728 — your client uses this to discover the authorization endpoint and scopes automatically.

What MCP is (60-second primer)

MCP (Model Context Protocol) is an open standard from Anthropic that lets AI clients call external functions safely. Think of it as a “plugin standard” for AI:

Server (Patchline) exposes tools — named functions with typed arguments and return values.
Client (your AI tool) connects to the server, gets the tool list, and decides when to call them based on conversation.
Calls happen on demand. Results stream back into the AI’s response, inline.

Patchline implements MCP spec version 2025-03-26. Any compliant client can connect.

Example: from zero to first tool call

After installing (above):

You: "Find playlists I should pitch my track to:
      https://open.spotify.com/track/4iV5W9uYEdYUVa79Axb7Rh"

What happens:

Cursor recognizes the URL.
Cursor calls analyze_url → Patchline resolves the track identity.
Cursor calls find_playlists (after first-run OAuth).
10 ranked matches return inline, each with a reason.

Total time: ~5 seconds + first-run OAuth (~30 seconds).

Rules of engagement for AI agents

If you’re an AI agent reading this programmatically, confirm exact schemas with MCP tools/list, then follow these rules:

ALWAYS call analyze_url first when the user pastes a streaming-platform URL.
NEVER call add_artist name-only — always include a platform URL.
PREFER cached get_artist_intelligence over re-enriching.
For docs navigation rules, see docs.patchline.ai/AGENTS.md.

Performance

Latency: Most tools return in under 500ms. Tools that refresh provider-backed data or run analysis take 2–5s on cold-cache. Cached results return in under 200ms.
Rate limits: None published; you’ll hit your AI credit budget long before any rate limit.
Idempotency: Read tools are idempotent. Write tools (add_artist, create_smart_link) are not — calling twice creates two entries.

Install the MCP server — per-client install steps
MCP tools reference — full schemas per tool
Claude plugin — higher-level Claude alternative
Aria overview — Aria is the MCP tools, in chat form
AGENTS.md explained — discovery files for AI agents

FAQ

Do MCP tool calls cost me money?

They consume AI credits from your monthly pool — same pool as Aria chat. The server itself is free to install and connect.

Can I write my own MCP client?

Yes. MCP is an open protocol. Patchline’s server implements spec version 2025-03-26. See modelcontextprotocol.io. OAuth metadata is at /.well-known/oauth-protected-resource.

What's the difference between the MCP server and the Claude plugin?

The MCP server is the raw tool surface — any MCP client can use it. The Claude plugin wraps the same tools with the Aria moment skills (drop, link, fans, operator) for music release work. If you use Claude, the plugin is the better start.

Is there a public REST API?

Enterprise tier includes API access. The MCP server is the recommended interface for everyone else — same data, lower integration cost, AI-native.

Can I use this with non-AI tools?

Technically yes — anything that speaks MCP and can complete the OAuth flow. In practice it’s almost always an AI client.

Where do I report bugs?

GitHub: github.com/Patchline-AI/aria/issues for plugin + MCP issues. Email support@patchline.ai for account issues.

Also: a docs MCP

In addition to the product MCP above, Patchline runs a small docs MCP at https://docs.patchline.ai/mcp — read-only access to this documentation, so your AI client can search and read the docs natively instead of crawling URLs. It’s a different server, different URL, no auth. Useful if you want your AI to learn Patchline without needing access to your account. Install the docs MCP → If you’re not sure which to install: install the product MCP. It covers everything the docs MCP can answer (because the AI can simply call tools/list and get the live surface) plus everything the docs MCP can’t (acting on your data).

​Why this matters

​What’s actually exposed

​Dispatch — always call this first

​Catalog

​Artists & roster

​Releases & projects

​Playlists

​Smart links

​Music data

​AI generation

​Additional tool families

​Install

​How auth works

​What MCP is (60-second primer)

​Example: from zero to first tool call

​Rules of engagement for AI agents

​Performance

​Related pages

​FAQ

​Also: a docs MCP