Enterprise DNA
M MCP Servers Developer low

gzoonet/cortex

by Various

Local-first knowledge graph for developers. Watches your files, builds a knowledge graph with LLMs, lets you query across projects.

G

MCP

gzoonet/cortex

Added 1 June 2026

#cli #developer-tools #gzoo #knowledge-graph #llm #local-first #mcp #ollama

Overview

A local-first knowledge graph for developers. It watches your files and uses large language models to build a graph of code relationships. You can then query across projects to find connections, dependencies, and contextual information.

Best for

Best for
Developers exploring cross-project code relationships with a preference for local data control

Use cases

  • Querying code relationships across multiple projects
  • Discovering dependencies and references between files
  • Exploring contextual connections without manual tagging

How to use

Install

npm install -g @gzoo/cortex

Tools exposed

  • cloud-first
  • local-first
  • local-only
  • cortex_ask
  • get_status
  • list_projects
  • find_entity
  • query_cortex
  • get_contradictions
  • resolve_contradiction
  • search_entities
  • ingest_file
  • add_project
  • remove_project
  • session_brief

Tested with

Claude Code, ChatGPT

Notes

A local-first knowledge graph for developers. It watches your files and uses large language models to build a graph of code relationships. You can then query across projects to find connections, dependencies, and contextual information.

15 stars on GitHub. Last updated 2026-05-25. Licensed MIT.

Use cases

  • Querying code relationships across multiple projects
  • Discovering dependencies and references between files
  • Exploring contextual connections without manual tagging

Pros

  • Local-first design keeps your code data on your machine
  • Automatically watches file changes to keep the graph updated
  • Cross-project queries enable holistic code understanding

Cons

  • Very early stage with only 15 GitHub stars, limited community and stability
  • Requires an LLM setup (local or API) which can be resource-intensive or costly
  • Knowledge graph quality depends on LLM performance and may be inconsistent

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.

Pros

  • Local-first design keeps your code data on your machine
  • Automatically watches file changes to keep the graph updated
  • Cross-project queries enable holistic code understanding

Cons

  • Very early stage with only 15 GitHub stars, limited community and stability
  • Requires an LLM setup (local or API) which can be resource-intensive or costly
  • Knowledge graph quality depends on LLM performance and may be inconsistent
Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.

Running a business, not writing the code? See the MCP servers picked for operators, and get your first one wired up with us.

Operator picks