$ loading_
Invoke Cursor SDK agents from MCP clients for coding and ongoing conversations.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "cursor-sdk-mcp" yet — see the docs or source repo.
Use cursor-sdk-mcp to invoke a Cursor SDK coding agent, analyze this TypeScript error, and return the fixed code with an explanation of changes.
Returns issue analysis, corrected code, and a summary of key changes.
Use cursor-sdk-mcp to list the models currently available in the Cursor SDK runtime, grouped by name.
Returns a list of available models for later agent selection.
Use cursor-sdk-mcp to continue a previous Cursor SDK conversation and finish the remaining refactoring tasks based on the prior discussion.
Continues with prior context and outputs follow-up guidance, code changes, or next steps.
Developers can invoke the Cursor SDK agent runtime directly from an MCP-enabled client without switching tools manually. It fits workflows that combine code analysis, edits, and conversation in one place.
When a team needs to select a model for different coding tasks, they can first list the models supported by Cursor SDK and then choose an agent. This makes model selection easier inside automated workflows.
If a coding task cannot be completed in one turn, users can continue an existing conversation and preserve context for later changes and decisions. This is useful for refactoring, debugging, and staged implementation.
It lets MCP clients invoke the Cursor SDK agent runtime, run coding agents, list models, and continue existing conversations. Its main purpose is to bring Cursor agent capabilities into an MCP environment.
Based on the description, it is mainly for coding-related tasks because it supports running coding agents. For broader task support, see the source repository.
The provided material does not include installation steps, runtime requirements, or key information. For prerequisites and setup details, see the source repository.
Delegate coding, shell tasks, and codebase queries to Cursor AI.
Manage Cursor agents with budget controls, monitoring, and inbox messaging.
Use Cursor MCP with a Google Docs agent for document automation and collaboration.
Compare AI models, estimate costs, and generate production-ready code in your IDE.
Build MCP servers once and connect them to multiple AI agents.
Connect MCP clients to basic tools and customizable prompts.