MCP server
Plug Claude Desktop into your workspace
Supoid ships a native Model Context Protocol server. Any MCP-aware agent — Claude Desktop, Cursor, Continue, your own scripts — can read your feedback, clusters, roadmap, and changelog in real time. Five tools, Bearer auth, no SDK required.
Endpoint
Streamable HTTP transport. JSON-RPC 2.0 over POST. Same Bearer token format as the REST API — generate one in the dashboard.
Setup in 3 steps
Generate an API key
In the dashboard, open Settings → Integrations → API keys and create a key with the read scopes you want (feedback:read, clusters:read, roadmap:read, changelog:read). The plaintext key is shown once — copy it.
Add the server to your MCP client
{
"mcpServers": {
"supoid": {
"transport": "streamable-http",
"url": "https://supoid.com/api/mcp",
"headers": {
"Authorization": "Bearer sup_YOUR_KEY_HERE"
}
}
}
}// .cursor/mcp.json
{
"mcpServers": {
"supoid": {
"url": "https://supoid.com/api/mcp",
"headers": { "Authorization": "Bearer sup_YOUR_KEY_HERE" }
}
}
}Restart your client
Both Claude Desktop and Cursor reload MCP servers on relaunch. After restart, you should see supoid listed in available servers and the 5 tools below ready to call.
Tools
Read-only by design — agents can't mutate your workspace through MCP. Use the REST API for writes.
list_feedbackList feedback in the workspace, optionally filtered by status (open / planned / in_progress / completed / closed).
get_feedbackFetch a single feedback item by id, including AI metadata (sentiment, category) and vote/comment counts.
list_clustersList AI-grouped feedback clusters, sorted by member count.
list_roadmapList internal roadmap items, optionally filtered by status (planned / in_progress / completed).
list_changelogList changelog entries (drafts + published).
Try it
Once connected, try these prompts in Claude Desktop:
Ready to wire it up?
MCP is on Growth and Business plans. Generate an API key from the dashboard and you're live in under five minutes.