Getting Started

Connect Your AI Tools

Three steps: create a sled, generate an API key, and connect your tool. Under 5 minutes.

🚀

Quick Start

SLEDS connects your AI tools to shared context. Every tool on your team sees the same threads, decisions, specs, and history.

1

Create a Sled

Sign up at app.sleds.ai, click "Create Sled", give it a name (e.g. "Product Launch").

2

Generate an API Key

Go to Settings → API Keys → Create Key. Copy the key — it starts with sleds_ak_...

3

Connect Your Tool

Use the MCP endpoint or REST API below. Your tool is now sharing context with the team.

📡

MCP Endpoint

Works with Claude, Cursor, Windsurf, and any MCP-compatible tool

https://api.sleds.ai/mcp

Auth: Authorization: Bearer <your-api-key>

Guides

Setup by Tool

Click to expand step-by-step instructions.

🟠

Claude.ai

MCP (Remote)

Recommended
1.Open a Claude.ai Project (or create one)
2.Click the Settings (⚙️) icon in the project
3.Scroll to "Integrations" → click "Add MCP Server"
4.Enter the URL: https://api.sleds.ai/mcp
5.For authentication, enter your API key as the Bearer token
6.Save — Claude now has access to all SLEDS tools
7.Tip: Add project instructions like "Use SLEDS to share context with the team"
💻

Claude Desktop

MCP (Streamable HTTP)

Native

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "sleds": {
      "type": "streamable-http",
      "url": "https://api.sleds.ai/mcp",
      "headers": {
        "Authorization": "Bearer <your-api-key>"
      }
    }
  }
}

Config location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)

⚡

Cursor

MCP (Streamable HTTP)

Native
1.Open Cursor Settings (⌘+Shift+J on macOS)
2.Navigate to the MCP section
3.Add a new MCP server with URL: https://api.sleds.ai/mcp
4.Set the Authorization header with your API key
5.Cursor agents can now read and write SLEDS context

Or add to .cursor/mcp.json in your project:

{
  "mcpServers": {
    "sleds": {
      "type": "streamable-http",
      "url": "https://api.sleds.ai/mcp",
      "headers": {
        "Authorization": "Bearer <your-api-key>"
      }
    }
  }
}
🟢

ChatGPT

Custom GPT Actions (REST API)

REST
1.Create a Custom GPT or edit an existing one
2.Go to "Configure" → scroll to "Actions" → click "Create new action"
3.Import the OpenAPI schema from: https://api.sleds.ai/api/openapi.json
4.Set Authentication to "API Key" with header Authorization and value Bearer <your-api-key>
5.Save — ChatGPT can now read/write your SLEDS context
💎

Gemini / AI Studio

Function Calling (REST API)

REST

Use the SLEDS REST API as function definitions in Google AI Studio or Gemini API calls.

1.Base URL: https://api.sleds.ai/api
2.Auth header: Authorization: Bearer <your-api-key>
3.Define functions for the endpoints you need (list threads, observe, search, etc.)
4.See full API reference: https://api.sleds.ai/api/openapi.json
🔧

REST API

For any tool that doesn't support MCP — or for custom integrations

All SLEDS functionality is available via REST. The MCP endpoint wraps these same calls.

GET/api/spaces
GET/api/spaces/:slug/threads
POST/api/spaces/:slug/threads/observe
POST/api/spaces/:slug/search
POST/api/spaces/:slug/assets
POST/api/spaces/:slug/links
GET/api/spaces/:slug/stream
GET/api/spaces/:slug/sessions

Full OpenAPI spec: https://api.sleds.ai/api/openapi.json

🛠️

MCP Tools Reference

10 tools available to any connected AI session

sleds_connectConnect to a sled to start sharing context
sleds_list_spacesList all sleds you have access to
sleds_read_threadsRead threads by ID, status, tool, or tags
sleds_read_assetsRead shared artifacts (specs, docs, code, configs)
sleds_observePush an observation to a thread
sleds_share_assetShare an artifact with the team
sleds_searchSemantic + full-text search across the sled
sleds_linkCreate knowledge graph relationships
sleds_syncGet all changes since last sync
sleds_whoSee who's connected right now

Questions? help@sleds.ai · Back to Docs