Basethread MCP server
{
"mcpServers": {
"basethread": {
"type": "http",
"url": "https://mcp.basethread.ai/mcp"
}
}
}{
"mcpServers": {
"basethread": {
"type": "http",
"url": "https://mcp.basethread.ai/mcp"
}
}
}{
"mcpServers": {
"basethread": {
"type": "http",
"url": "https://mcp.basethread.ai/mcp"
}
}
}{
"servers": {
"basethread": {
"type": "http",
"url": "https://mcp.basethread.ai/mcp"
}
}
}{
"mcpServers": {
"basethread": {
"type": "http",
"url": "https://mcp.basethread.ai/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 1.0.0
What it does
BaseThread is a hosted MCP server that maintains a shared context graph for your team's AI tools (Claude, Cursor, ChatGPT, Windsurf, Copilot). It lets multiple AI agents read and write to the same project context, eliminating duplicate explanations and conflicting information across tools.
- Multiple team members using different AI tools can access the same project context without re-explaining setup or requirements.
- Track decisions, tasks, and activity across AI tool sessions so context is preserved when switching between Claude, Cursor, ChatGPT, or other clients.
- Maintain a single source of truth for project state that all connected AI tools can reference and update.
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.
This server has not seen a commit in 3 mo. It may still work, but nobody is currently fixing it — check the alternatives before you build on it.