Unified MCP server: hybrid search + code graph + semantic memory. 15 tools, <350MB RAM, fully local. No API keys.
Quick Install
Copy the config for your editor. Some servers may need additional setup — check the README.
Claude Desktop
Claude Code
Cursor
Add to claude_desktop_config.json:
{
"mcpServers": {
"nexus-mcp": {
"command": "uvx",
"args": [
"nexus-mcp"
]
}
}
}
📋 Copy
Run in terminal:
claude mcp add nexus-mcp uvx nexus-mcp
📋 Copy
Add to .cursor/mcp.json:
{
"mcpServers": {
"nexus-mcp": {
"command": "uvx",
"args": [
"nexus-mcp"
]
}
}
}
📋 Copy
Or install with pip: pip install nexus-mcp
README Excerpt
**Hybrid search + code graph + semantic memory in a single local MCP server — under 350 MB RAM.** Nexus-MCP is a code intelligence server for the [Model Context Protocol](https://modelcontextprotocol.io). It gives AI agents precise, token-efficient answers about your codebase without cloud dependencies: no API keys, no data egress, no subscriptions.
Tools (20)
NEXUS_AUDIT_ENABLED NEXUS_CHUNK_MAX_CHARS NEXUS_EMBEDDING_DEVICE NEXUS_EMBEDDING_MODEL NEXUS_FUSION_WEIGHT_BM25 NEXUS_FUSION_WEIGHT_GRAPH NEXUS_FUSION_WEIGHT_VECTOR NEXUS_LOG_FORMAT NEXUS_LOG_LEVEL NEXUS_MAX_FILE_SIZE_MB NEXUS_MAX_MEMORY_MB NEXUS_PERMISSION_LEVEL NEXUS_RATE_LIMIT_ENABLED NEXUS_SEARCH_MODE NEXUS_STORAGE_DIR NEXUS_TRUST_REMOTE_CODE detailed full mode summary
Topics
ai ai-agent ai-coding claude code-analysis code-graph code-search hybrid-search lancedb llm local-first mcp mcp-server model-context-protocol onnx