shantanugoel

shantanugoel / oxydra

Public

oxydra - An agent with many heads, zero overhead

35
3
69% credibility
Found Feb 27, 2026 at 22 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Oxydra is a Rust-based orchestrator for building and running AI agents with multi-provider LLM support, sandboxed tools, persistent memory, task scheduling, and chat interfaces like a terminal UI and Telegram bots.

How It Works

1
🔍 Discover Oxydra

You hear about Oxydra, a smart helper that runs AI agents to handle tasks like chatting, remembering details, and scheduling work.

2
📋 Gather setup pieces

You copy a few simple guide files into a folder to prepare your personal AI workspace.

3
🧠 Connect your AI brain

You link a thinking service like Claude or ChatGPT so your agents can reason and respond.

4
🛡️ Prepare safe workspaces

You create secure play areas where AI can safely use tools without risks.

5
🚀 Launch your assistant

With one command, you start your AI companion, ready to listen on screen or messages.

6
Chat however you like
🖥️
Screen chat

Type messages in a friendly terminal window, switch conversations easily.

📱
Messaging app

Send texts, photos, or files via Telegram bot for on-the-go help.

🎉 AI takes over tasks

Your agents remember past chats, run scheduled jobs, and deliver results right where you need them, making life easier.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 22 to 35 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 oxydra?

Oxydra is a Rust-based AI agent orchestrator that runs multi-headed agents with zero overhead, handling provider-agnostic LLMs like OpenAI, Anthropic, and Gemini. It solves the pain of coordinating complex AI workflows by offering persistent memory via libSQL, tool execution with safety tiers, task scheduling, and isolation in containers or MicroVMs. Users get a WebSocket gateway for multi-session management, a TUI for interaction, and Telegram integration for multi-modal inputs like images or PDFs.

Why is it gaining traction?

Its Rust core delivers high performance without the bloat of Python frameworks, standing out with seamless streaming responses, automatic retries, and cron-based scheduling that routes back to origins like Telegram topics. Developers hook on the proc-macro for tools, hybrid vector/FTS memory search, and per-user configs—making agentic apps feel lightweight yet production-ready. Zero-overhead multi-agent coordination via turn-loop state machines keeps costs low.

Who should use this?

Rust enthusiasts prototyping AI agents for Telegram bots or scheduled tasks, like research automation in forum threads. Teams building secure, isolated workflows with file/vault tools and LLM delegation, especially those ditching heavy orchestrators for something lean. Early adopters needing TUI-driven multi-session debugging.

Verdict

Skip for now—18 stars and pre-pre-alpha status scream WIP, with a 0.699999988079071% credibility score signaling high risk outside tinkering. Promising for Rust AI devs willing to build guests and configs, but wait for stability unless you're the author.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.