Skip to main content
The SureFeedback MCP Server lets AI assistants like Claude Code, Cursor, and Windsurf read and act on your feedback data directly — no copy-pasting, no context switching.

10 Tools

Search comments, summarize feedback, get pending tasks, update statuses, assign teammates, and more.

7 Resources

Read comments, sites, mockups, and members as structured data using surefeedback:// URIs.

Project-scoped

Every MCP key is scoped to your active project. No project IDs needed in queries.

JSON-RPC 2.0

Standard Streamable HTTP transport. Works with any MCP-compatible client.

​
How it works

AI Assistant  →  MCP Client  →  POST https://api.surefeedback.com/mcp  →  SureFeedback
Your AI assistant calls tools like search-comments or get-pending-tasks. The MCP server authenticates via your API key, queries your project data, and returns structured results the AI can reason over.

​
Example prompts

Once connected, you can ask your AI assistant:
  • “Summarize all feedback for this project”
  • “Show me all pending tasks sorted by priority”
  • “Find all urgent open comments on the homepage”
  • “Assign comment SF-042 to [email protected]”
  • “Mark comment SF-017 as resolved”
  • “List all members in this organisation”

​
Transport

The SureFeedback MCP server uses Streamable HTTP transport over a single POST endpoint: 
POST https://api.surefeedback.com/mcp
All MCP operations — initialize, tools/list, tools/call, resources/read — are sent as JSON-RPC 2.0 messages to this endpoint.