lattice-mcp

mcp-serverNo score yet

MCP server for token-efficient large document analysis via the use of REPL state

Stars

140

Δ stars 7d

Δ stars 30d

Forks

16

Contributors

5

npm DL / wk

PyPI DL / wk

Language

TypeScript

Last push

2026-05-17

About lattice-mcp

Process documents 100x larger than your LLM's context window—without vector databases or chunking heuristics. LLMs have fixed context windows. Traditional solutions (RAG, chunking) lose information or miss connections across chunks. RLM takes a different approach: the model reasons about your query and outputs symbolic commands that a logic engine executes against the document. Unlike traditional approaches where an LLM writes arbitrary code, RLM uses **Nucleus**—a constrained symbolic language based on S-expressions. The LLM outputs Nucleus…

Read the full README on GitHub →

Newly tracked — a growth score appears after the first full scoring week. Signals are deliberately suppressed during cold start to avoid false positives.

lattice-mcp alternatives

Projects in the same category, closest in size — picked by data, not opinion.

See all mcp-server projects ranked by growth →

Frequently asked questions

Is lattice-mcp still maintained?
Activity is low: the last push was on 2026-05-17 (32 days before our latest snapshot).
What are the best lattice-mcp alternatives?
Closest by category and size in our data: reMarkable MCP Server, io.github.sathergate/clocktower, doggybee/mcp-server-ccxt — full list with live signals above.

Topics

mcpmcp-serverllmmodel-context-protocolllm-toolsai-assistantdocument-analysis

Embed this badge

Show your project's live signal in your README — it updates weekly with the data.

[![lattice-mcp on What's Trending in AI](https://whatstrendinginai.com/badges/yogthos-matryoshka-mcp-server.svg)](https://whatstrendinginai.com/projects/yogthos-matryoshka-mcp-server/)

lattice-mcp badge

Tracked since 2026-06-17 · data as of 2026-06-18 · 0 open issues ·