Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.spote.cloud/llms.txt

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

Spote includes a built-in Model Context Protocol (MCP) server that lets any compatible AI agent interact with your notes. Agents can save notes, search them, and retrieve past context — without switching apps.

​
Endpoints

Spote exposes two transport endpoints depending on what your AI client supports:
TransportURL
SSE (Server-Sent Events)https://your-spote-app.com/sse
HTTP streaminghttps://your-spote-app.com/mcp
Most modern MCP clients default to HTTP streaming (/mcp). Use /sse for clients that require SSE transport.

​
Authentication

Agents authenticate using a Personal Access Token (PAT) as a Bearer token in the Authorization header: 
Authorization: Bearer spote_<your-token>
You can generate a PAT from Settings in the Spote web app. See Personal Access Tokens for step-by-step instructions.

​
What agents can do

Save notes

Tell your agent to save a note with a title, content, bucket, and tags. Spote indexes it immediately for semantic search.

Search semantically

Ask your agent to find notes about a topic. Spote searches by meaning, not exact wording — so you find notes even when you don’t remember the phrasing you used.

List & retrieve

List your most recent notes or fetch the full content of a specific note by ID. Useful for pulling context into a conversation.

Manage notes

Update or delete notes from within a conversation. Your agent can reorganize, retitle, or remove notes without you opening the app.

​
Supported AI tools

The Spote MCP server works with any client that supports the MCP protocol over SSE or HTTP streaming transport, including:
  • Claude Desktop (Anthropic)
  • Cursor
  • Windsurf
  • Any tool that implements the MCP protocol with SSE or HTTP transport

​
Next steps

Connect an AI agent

Step-by-step setup for Claude Desktop, Cursor, and other MCP clients.

MCP tools reference

Full reference for every tool your agent can call, with parameters and return values.