Cortex Memory Plugin

v1.0.0Knowledge & Memorystable

Cortex is a Claude Code plugin that gives Claude persistent memory across sessions. A background miner extracts durable lessons (decisions, conventions, bug fixes, preferences) from your completed sessions, stores them as markdown in an Obsidian vaul

agent-memory-systemanthropicartificial-intelligencecausal-inferenceclaude
Share:
37
Stars
0
Downloads
0
Weekly
0/5

What is Cortex Memory Plugin?

Cortex Memory Plugin is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to cortex is a claude code plugin that gives claude persistent memory across sessions. a background miner extracts durable lessons (decisions, conventions, bug fixes, preferences) from your completed ses...

Cortex is a Claude Code plugin that gives Claude persistent memory across sessions. A background miner extracts durable lessons (decisions, conventions, bug fixes, preferences) from your completed sessions, stores them as markdown in an Obsidian vaul

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Cortex is a Claude Code plugin that gives Claude persistent

Use Cases

Claude Code persistent memory
Session-to-session knowledge
Decision extraction
cdeust

Maintainer

LicenseNOASSERTION
Languagepython
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx cortex

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Cortex Memory Plugin

Cortex is a Claude Code plugin that gives Claude persistent memory across sessions by running a background process that extracts durable lessons — architectural decisions, coding conventions, bug fixes, and preferences — from your completed sessions and stores them as structured markdown in a PostgreSQL-backed knowledge base. It provides 46 MCP tools covering memory recall, wiki authoring, codebase analysis, and context assembly, using five fused retrieval signals (vector similarity, full-text, trigram, thermodynamic heat, and recency) to surface the most relevant past knowledge. Teams and solo developers use it to prevent knowledge loss between sessions and build a growing, queryable record of project understanding.

Prerequisites

  • Claude Code CLI installed (the plugin runs inside Claude Code, not standalone Claude Desktop)
  • PostgreSQL instance accessible at localhost:5432 or via a custom DATABASE_URL
  • Python 3.10 or later (the MCP server is Python-based)
  • Sufficient disk space for an embedding model download during /cortex-setup-project
1

Install Cortex via Claude Code marketplace

Install Cortex using the Claude Code plugin marketplace commands. Do not use the PyPI package — it is deprecated and contains security vulnerabilities.

claude plugin marketplace add cdeust/Cortex
claude plugin install cortex
2

Restart Claude Code

After installation, restart Claude Code to load the Cortex plugin and make its slash commands available.

3

Run the project setup command

Inside Claude Code, run the setup command. It will automatically set up PostgreSQL, create the database, download an embedding model, and initialize the Cortex knowledge base for your project.

/cortex-setup-project
4

Configure environment variables if needed

Cortex works with default settings out of the box. If you need a custom PostgreSQL connection or want to tune behavior, set these environment variables.

export DATABASE_URL=postgresql://localhost:5432/cortex
export CORTEX_CONSOLIDATE_TTL_HOURS=6
export CORTEX_MCP_CALL_TIMEOUT_S=600
5

Verify the installation

Run the Cortex doctor script to confirm all components are healthy.

python3 -m mcp_server.doctor

Cortex Memory Plugin Examples

Client configuration

Cortex is installed as a Claude Code plugin and configured automatically. The MCP server environment can be tuned with these variables in your shell or .env file.

{
  "mcpServers": {
    "cortex": {
      "command": "python3",
      "args": ["-m", "mcp_server"],
      "env": {
        "DATABASE_URL": "postgresql://localhost:5432/cortex",
        "CORTEX_CONSOLIDATE_TTL_HOURS": "6",
        "CORTEX_MCP_CALL_TIMEOUT_S": "600"
      }
    }
  }
}

Prompts to try

After setup, use these prompts inside Claude Code sessions to leverage Cortex memory.

- "What architectural decisions have we made in this project?"
- "Recall any past bugs we fixed related to authentication"
- "Store this decision: we are using event sourcing for all state changes"
- "What do you remember about how we handle database migrations?"
- "Summarize what you know about the payment integration we built"

Troubleshooting Cortex Memory Plugin

/cortex-setup-project fails with PostgreSQL connection error

Ensure PostgreSQL is running locally on port 5432 and your user has permission to create databases. If using a remote PostgreSQL instance, set DATABASE_URL to the correct connection string before running setup.

Memory recall returns stale or irrelevant results

Run the background consolidation process manually. The CORTEX_CONSOLIDATE_TTL_HOURS variable controls how often automatic consolidation runs (default 6 hours). Reduce this value for more frequent knowledge updates.

Plugin not found after installation

Fully restart Claude Code after running claude plugin install cortex. The plugin system requires a fresh process start to discover and register new plugins.

Frequently Asked Questions about Cortex Memory Plugin

What is Cortex Memory Plugin?

Cortex Memory Plugin is a Model Context Protocol (MCP) server that cortex is a claude code plugin that gives claude persistent memory across sessions. a background miner extracts durable lessons (decisions, conventions, bug fixes, preferences) from your completed sessions, stores them as markdown in an obsidian vaul It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Cortex Memory Plugin?

Follow the installation instructions on the Cortex Memory Plugin GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Cortex Memory Plugin?

Cortex Memory Plugin works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Cortex Memory Plugin free to use?

Yes, Cortex Memory Plugin is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "cortex": { "command": "npx", "args": ["-y", "cortex"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Cortex Memory Plugin?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides