cortex
by dinglebear-ai
Cortex is a self-hosted log collection and search tool for home servers and small private networks. It gathers syslog, Docker, OpenTelemetry and AI assistant transcripts into a SQLite database, which you can search from a command line, a web page or an AI assistant.
- Stars
- 4
- Licence
- AGPL-3.0
- Runs on
- A small server
- Setup
- Moderate
- Language
- Rust
- Age
- 220 days
What you can do with it
- Collect syslog, Docker logs and OpenTelemetry data from several machines into one searchable database
- Ask an AI assistant to search logs, build timelines and investigate errors through its MCP connection
- Index Claude, Codex and Gemini CLI session transcripts and link them to system events
Make money with it
A freelancer could offer managed monitoring and troubleshooting for small clients' servers, using Cortex as the log back end.
An idea, not financial advice. Check the licence before charging for anything built on it.
Ways a business could use it
- Keep logs from a small office server or a few client machines in one place, without paying for a hosted logging service
- Speed up fault-finding on self-managed servers by searching errors across sources and having an AI assistant summarise incidents
Can you use it commercially?
Free to use. Anything you distribute that includes it must be open-sourced under the same licence. Running it as an online service also counts, so a hosted version must offer its source.
Not legal advice. Read the project's licence file before building a product on it.
Add it to Claude Code
claude mcp add cortex -- npx -y @dinglebear/cortex
From the official MCP Registry (ai.dinglebear/cortex). Only connect MCP servers you trust: they can act on your behalf.
Getting it running
Install the command-line tool with npm (Node.js 18 or newer) or build it from source with Rust, then run "cortex serve mcp" after setting a database path and access tokens.
Follow the official instructions in the project README.
Things to check
- Very new project with only a few stars
- Licensed under AGPL-3.0, which has obligations if you offer it as a network service
- Designed for trusted private networks, not for exposure to the public internet
- Not a clustered or large-scale logging system
Checked 03 Nov 2026. Summary drafted by AI from the project's README.
