Cortex
by cdeust·★ 34·Score 47
Cortex is a neuroscience-based persistent memory system for Claude Code with 47 MCP tools that remembers context, decisions, and patterns across sessions.
Overview
Cortex is a sophisticated memory engine for Claude Code built on computational neuroscience principles. It creates a persistent memory system that consolidates, forgets intelligently, and reconstructs the right context at the right time. The system uses 26 biological mechanisms, PostgreSQL with pgvector, and includes 47 MCP tools with 9 automatic hooks. It goes beyond simple text retrieval to understand the relationships between code elements, using a neural graph to represent projects, files, functions, and their connections. The system has been thoroughly tested with benchmarks showing 98.4% Recall@10 and 0.9124 MRR on LongMemEval.
Try asking AI
After installing, here are 5 things you can ask your AI assistant:
When to choose this
Choose Cortex when you need long-term memory for code development across multiple sessions, especially when working on complex systems where context persistence significantly improves productivity.
When NOT to choose this
Avoid Cortex if you need a lightweight solution without PostgreSQL dependency, or if you're working on short-term projects where context persistence doesn't provide significant value.
Tools this server exposes
1 tool extracted from the README (low confidence)query_workflow_graphReturns typed subgraphs from the workflow graph, filterable by node kind, edge kind, or relationship to a specific node
Note: The README mentions '47 MCP tools' but only explicitly names one (query_workflow_graph) in the text. I'm extracting only this one as it's explicitly mentioned. The server is described as having many tools, but their names and purposes aren'
Comparable tools
Installation
claude plugin marketplace add cdeust/Cortex
claude plugin install cortexRestart your Claude Code session, then run:
/cortex-setup-projectThis handles PostgreSQL + pgvector installation, database creation, embedding model download, cognitive profile building, codebase seeding, conversation import, and hook registration.
Verify setup:
python3 -m mcp_server.doctorFAQ
- What databases does Cortex use?
- Cortex uses PostgreSQL with pgvector for storing embeddings and creating vector searches.
- Can Cortex work with different programming languages?
- Yes, Cortex can parse and work with code in 10 languages including Rust, Python, TypeScript, Java, Kotlin, Swift, Objective-C, C, C++, and Go.
Compare Cortex with
Last updated · Auto-generated from public README + GitHub signals.