Stars
Make any agent harness multimodal-native.
🔍 Search anyone's digital footprint across 300+ websites
Official implementation of Continual Harness (arxiv.org/abs/2605.09998) on ARC-AGI-3
A self-improving RLM agent for coding workflows and long-running autonomous tasks.
Distributed 56M-parameter LLM inference across 3 ESP32-S3 boards via ESP-NOW , Split-PLE + KV cache, fully offline.
Rebuild the object in a reference image as a code-only, procedural, quality-gated, animation-ready Three.js model. Token-efficient image-to-3D.
Implementation of Googles Generative Anthropometric Model (Currently Head only) into a Blender Add-on.
An open ecosystem of parametric human models and perception stacks, starting with GNM Head.
Zenith: a continuous-improvement harness for long-running agent tasks. Turns Claude Code, Codex, or Hermes into a multi-agent mission orchestrator via MCP/ACP.
Infinite Worlds with Versatile Interactions
A multi-instrument music transcription model developed by Kyutai and Mirelo.
The open-source skills that turn your agent into a full PR team.
Miso TTS is an 8 billion, highly emotive text-to-speech model
Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and supply-chain risks in Claude Code, Codex, and MCP skills b…
alternative to claude -p that lifecycles interactive sessions in tmux
Use Codex from Claude Code to review code or delegate tasks.
A Chrome extension that fixes any image upload box. Reads upload constraints, crops, creates, or draws to spec.
A universal skill for generating realistic images
Reference code for the Meta-Harness paper.
A calm, CLI-native way to semantically grep everything, like code, images, pdfs and more.
Official Codebase for CVPR 2026 Highlight Paper: "Building a Precise Video Language with Human–AI Oversight"