by-scott

by-scott / cortex

Public

Cognitive runtime for language models with memory, metacognition, multimodal channels, native plugins, and a self-evolving Executive.

47
1
100% credibility
Found Apr 26, 2026 at 47 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Cortex is a local runtime that turns language models into persistent, multi-interface AI agents with memory, tools, metacognition, and plugin support.

How It Works

1
💡 Discover a smart AI companion

You hear about Cortex, a helpful AI that remembers conversations, uses tools, and works across your apps like a personal assistant.

2
🔧 Set it up on your computer

Run a simple command to install it locally, like adding a new app to your machine.

3
🧠 Connect your AI service

Link it to your favorite AI provider so it can think and respond intelligently.

4
💬 Start chatting right away

Open the chat window, talk naturally, and watch it remember details across sessions.

5
Connect to your apps
📲
Messaging apps

Pair with Telegram, WhatsApp, or QQ for on-the-go conversations.

🌐
Web dashboard

Access via browser for a full view of memory and tasks.

6
🛠️ Add helpful extensions

Install plugins like a coding helper to make it even smarter for your needs.

🚀 Your AI lives and grows

Now it remembers everything, acts on its own when needed, and helps across all your devices seamlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 47 to 47 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is cortex?

Cortex is a Rust-built cognitive runtime that turns language models into persistent, local agents with durable memory, metacognition, and multimodal channels for text, vision, audio, and video. It runs as a daemon handling CLI REPL, HTTP APIs, JSON-RPC, and messaging integrations like Telegram, WhatsApp, and QQ, while journaling turns for replay and enforcing risk gates on tools. Drawing from cognitive science github ideas, it solves agent drift and cognitive overload github by sustaining goal-directed behavior across sessions.

Why is it gaining traction?

It stands out from agent frameworks by treating cognition as an OS substrate: explicit turn states, self-tuning metacognitive alerts, and a repertoire of evolving skills that adapt via usage stats, not just prompts. Hot-reload plugins add tools/media/skills without restarts, and cross-transport identity keeps your agent consistent whether via CLI or chat apps. For cognitive ai github enthusiasts, the Rust safety and cognitive architecture github primitives like attention scheduling hook devs building reliable local LLMs.

Who should use this?

AI tinkerers prototyping cognitive tools github with memory and self-reflection, solo devs turning it into a coding agent via the dev plugin (git ops, tree-sitter parsing, multi-agent teams), or ops folks needing a cortex github llm daemon with permission modes for secure tool calls. Skip if you want quick cloud spins; grab it for local, multi-interface persistence.

Verdict

Early days at 47 stars and 1.0% credibility—maturity lags with big features like event-sourcing unproven in prod—but docs and quickstart shine for Rust fans. Prototype now if cognitive runtime appeals; production wait for hardening.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.