Skip to main content
Glama
orneryd

M.I.M.I.R - Multi-agent Intelligent Memory & Insight Repository

by orneryd
MCP_CURSOR_TEST.md3.91 kB
# MCP Server Test in Cursor This document contains test prompts to verify the MCP server is working correctly in Cursor. --- ## Test 1: Create a TODO Node **Prompt for Cursor:** ``` Create a TODO using the memory_add_node tool with these properties: - type: "todo" - description: "Test MCP connection from Cursor" - status: "pending" - priority: "high" ``` **Expected Result:** - Should return a node with an ID like `todo-1-xxxxx` - Should show all properties - Should include created/updated timestamps --- ## Test 2: Query All TODO Nodes **Prompt for Cursor:** ``` Query all TODO nodes using memory_query_nodes with type "todo" ``` **Expected Result:** - Should return a list of all TODO nodes - Should include the one we just created --- ## Test 3: Create Multiple Nodes with Edges **Prompt for Cursor:** ``` Create a project structure: 1. Create a project node (type: "project", name: "MCP Testing", status: "active") 2. Create a file node (type: "file", path: "test.ts", language: "typescript") 3. Create an edge connecting the project to the file (type: "contains") ``` **Expected Result:** - Should create 2 nodes and 1 edge - Should return IDs for all created items --- ## Test 4: Get Subgraph **Prompt for Cursor:** ``` Get the subgraph for the project node we just created with depth 2 ``` **Expected Result:** - Should return the project node - Should return connected file node - Should return the edge between them --- ## Test 5: Search Nodes **Prompt for Cursor:** ``` Search for nodes containing "MCP" using memory_search_nodes ``` **Expected Result:** - Should return nodes matching "MCP" in their properties - Should use full-text search --- ## Test 6: Update a Node **Prompt for Cursor:** ``` Update the first TODO we created - change status to "in_progress" ``` **Expected Result:** - Should successfully update the status - Should show updated timestamp changed - Should preserve other properties --- ## Test 7: Get Node with Neighbors **Prompt for Cursor:** ``` Get all neighbors of the project node ``` **Expected Result:** - Should return the file node - Should show relationship information --- ## Test 8: Batch Operations **Prompt for Cursor:** ``` Create 3 TODO nodes in a single batch operation: 1. "Implement feature A", priority: high, status: pending 2. "Review code", priority: medium, status: pending 3. "Write tests", priority: high, status: pending ``` **Expected Result:** - Should create all 3 nodes in one operation - Should return array of created nodes --- ## Test 9: Get Graph Statistics **Prompt for Cursor:** ``` What are the current graph statistics? How many nodes and edges do we have? ``` **Note:** This might require calling memory_query_nodes without filters or checking multiple types. --- ## Test 10: Clean Up **Prompt for Cursor:** ``` Delete all the test nodes we created ``` **Warning:** Only do this after all other tests are complete! --- ## Debugging Tips If a tool call fails: 1. Check Docker is running: `docker-compose ps` 2. Check MCP server logs: `docker logs mcp_server -f` 3. Check Neo4j is healthy: `docker exec neo4j_db cypher-shell -u neo4j -p password "RETURN 1"` 4. Verify health endpoint: `curl http://localhost:3000/health` --- ## Expected Tools Available Cursor should show these 17 tools: **Single Operations:** - memory_add_node - memory_get_node - memory_update_node - memory_delete_node - memory_add_edge - memory_delete_edge - memory_query_nodes - memory_search_nodes - memory_get_edges - memory_get_neighbors - memory_get_subgraph - memory_clear **Batch Operations:** - memory_add_nodes - memory_update_nodes - memory_delete_nodes - memory_add_edges - memory_delete_edges --- ## Success Criteria ✅ All tool calls complete without errors ✅ Data persists across calls ✅ Relationships work correctly ✅ Search returns relevant results ✅ Batch operations work efficiently

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/orneryd/Mimir'

If you have feedback or need assistance with the MCP directory API, please join our Discord server