# Test MCP Codebase Index
## How to Test
1. **Check if server is running**: Look for "Connection state: Running" in logs ✅
2. **Test search in VS Code**:
- Open Copilot Chat
- Use the `@codebase` mention (if configured)
- Or ask: "Search the codebase for authentication code"
3. **Expected behavior**:
- Server processes the query
- Returns relevant code chunks from your project
- You see results in chat
## What the warnings mean
```
[warning] Failed to parse message: "[Embedder] Processed 2/2 chunks\n"
```
**This is NORMAL** - These are info logs from the server that VS Code can't parse as MCP protocol messages. They don't indicate errors.
## Real errors to look for
❌ **These are actual errors:**
```
[error] Request payload size exceeds the limit: 36000 bytes
[error] Failed to embed /path/to/file
[error] Connection refused
[error] GEMINI_API_KEY not found
```
✅ **Current status:**
- ✅ Connection state: Running
- ✅ Loaded 1472 vectors
- ✅ Processing files without size limit errors
- ✅ No "exceeds the limit" messages
## Troubleshooting
If search doesn't work:
1. **Wait for indexing to complete** - Check logs for "[Indexer] Indexing complete"
2. **Verify tool is available** - In Copilot Chat, type `@` and see if codebase search appears
3. **Check config** - Ensure `REPO_PATH` and `GEMINI_API_KEY` are set correctly
## Test queries
Try these in Copilot Chat:
```
"Find authentication logic in the codebase"
"Show me how database connections are configured"
"Where is error handling implemented?"
"Find all API endpoint definitions"
```
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/NgoTaiCo/mcp-codebase-index'
If you have feedback or need assistance with the MCP directory API, please join our Discord server