general MCP Server
MCP server for token-efficient large document analysis via the use of REPL state
Discovered via github-seeds and last synced 3mo ago.
1. Install the package
npx matryoshka-rlm
2. Add to claude_desktop_config.json
{
"mcpServers": {
"matryoshka-rlm": {
"command": "npx",
"args": [
"matryoshka-rlm"
]
}
}
}Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)
Purpose
Description
Required
Yes
No
Yes
Yes
Show current variable bindings and memo labels
Reset all bindings and memos but keep document loaded
Get Nucleus command reference
Description
Main CLI for document analysis with LLM reasoning
Execute Nucleus commands on the loaded document
MCP server exposing direct Nucleus commands (no LLM required)
Store arbitrary context as a memo handle (no document required)
MCP server with full RLM + LLM orchestration (`analyze_document` tool)
Respond to a pending `(llm_query ...)` suspension
`runRLM` (CLI / programmatic)
Pipe adapter for programmatic access
Close the session and free memory
HTTP server for Nucleus queries
Delete a stale memo to free memory
Load a document for analysis
✅
Setup script for Claude Code integration
Get session status, document info, and memo usage
Expand a handle to see full data (with optional limit/offset)
Respond to a pending `(llm_batch ...)` suspension with all N responses
Interactive REPL for Nucleus commands