Hierarchical markdown memory palace for AI agents — structured palace navigation via MCP tools.
Hierarchical markdown memory palace for AI agents — structured palace navigation via MCP tools.
Is it safe?
No known CVEs for locus-mcp.
No authentication — any process on your machine can connect to this server.
License not specified.
Last scanned 0 days ago.
Is it maintained?
Last commit 20 days ago.
Will it work with my client?
Transport: stdio, sse, http. Works with Claude Desktop, Cursor, Claude Code, and most MCP clients.
How much context will it use?
5 tools. Estimated ~200 tokens of your context window (0.1% of 200K).
What if it doesn't work?
Common issues: JSON syntax errors in config, Python version mismatch, network or firewall blocking. covers troubleshooting.
{
"mcpServers": {
"locus": {
"args": [
"--palace",
"/path/to/palace"
],
"command": "locus-mcp"
}
}
}Run this in your terminal to verify the server starts. Then let us know if it worked — your result helps other developers.
uvx locus-mcp 2>&1 | head -1 && echo "✓ Server started successfully"
After testing, let us know if it worked:
memory_listReturns INDEX.md (no args) or lists a room's files
memory_readReads any file in the palace
memory_writeAtomically writes a file (guarded — cannot write to _metrics/, sessions/, .sig/, .security/)
memory_searchFull-text search across the palace (ripgrep or Python fallback)
memory_batchReads up to 20 palace files in a single call — use for multi-room loads
Last scanned 6h ago
No known vulnerabilities.
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.