Skip to content
View osolmaz's full-sized avatar
πŸ«–
πŸ«–

Highlights

  • Pro

Organizations

@huggingface @horsefit @openclaw @dutifuldev

Block or report osolmaz

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
osolmaz/README.md

Hi there πŸ‘‹

It's Onur. I work at Hugging Face πŸ€—, where my current focus is making OpenClaw 🦞 and other agent harnesses work great with local models.

I am addicted to agents and I have too many side projects. You might get an idea of me if you follow me on X:onusoz or Hugging Face: osolmaz.

Personal/closed source projects

  • 🧠 TextCortex: Created TextCortex's AI model harness, starting out as a hobby project a few months before ChatGPT launched
  • 🐴 Horse.fit: AI personal trainer (currently counts push-ups and lets you challenge your friends
  • 😎 solmaz.io: My blog

Agents

  • 🫑 Bob: My personal agent, helping me out in many tasks

Open source projects (actively maintained):

  • πŸ“Ÿ acpx: Headless CLI client for statful ACP sessions
  • 🦞 OpenClaw: I'm a maintainer of various features in the biggest personal agent project
  • 🍨 Scoop: Open personal news intelligence platform for OpenClaw
  • πŸ“„ JSON-DOC: Open block-based document file format based on Notion
  • 🍹 Spritz: Kubernetes‑native control plane for running AI agents in containers
  • 🧹 Janitr: Smol local models to filter out internet content for you
  • πŸ₯” Alman.AI: Alman, a Simplified Dialect of the German Language
  • ✨ SimpleDoc: Lightweight standard for organizing Markdown documentation in codebases
  • ⛳️ skillflag: Simple CLI flag convention for listing and installing agent skills
  • 🧰 localpi: Swiss army knife for running Pi with local inference engines such as llama.cpp, vLLM, SGLang, LM Studio, Ollama, etc.
  • 🏭 pi-factory: Declarative Pi application bundles
  • πŸ“ˆ localperf: Local inference performance benchmarking and profiling
  • 🧭 ghzinga: Clickable terminal UI for viewing a GitHub issue or pull request
  • πŸ”¨ Slophammer: Quality-gate and reference tooling for coding projects
  • 🧬 Schemator: Schema and data-model review CLI for agents
  • πŸ›‘οΈ GitHub Sane Defaults: Applies sane GitHub organization and repository defaults
  • πŸͺž ghreplica: Local GitHub mirror server to avoid rate limits
  • πŸ—ƒοΈ xTap Sync: Syncs xTap JSONL exports into a Git-backed tweet archive
  • πŸ”— Claude MD Symlinker: Keeps AGENTS.md and CLAUDE.md in sync with symlinks
  • 🚨 Localpager: Local-first triage and paging tool for GitHub issues and pull requests
  • πŸ€— ML Claw: Deploy OpenClaw agents on Hugging Face

Open source projects (unmaintained):

  • 🎀 Manim Voiceover: Plugin for programmatic voiceovers in Manim videos
  • 🎬 Manim Euclid Elements: Video rendering of Euclid's Elements using Manim
  • 🧠 icortex: A Jupyter notebook coding agent that I created 2 months before ChatGPT existed, using code-davinci-002
  • πŸ“¦ Claude Code Sandbox: A vibe-coded PoC for running Claude Code locally in a sandbox, and controlling through a browser UI

This list is probably not up-to-date, make sure to follow me on X for the latest info

Pinned Loading

  1. pi-workflows pi-workflows Public

    Workflow engine, JSON control-flow tool, and live terminal viewer for the pi coding agent

    TypeScript 246 14

  2. openclaw/acpx openclaw/acpx Public

    Headless CLI client for stateful Agent Client Protocol (ACP) sessions

    TypeScript 3.2k 329

  3. ManimCommunity/manim-voiceover ManimCommunity/manim-voiceover Public

    Manim plugin for all things voiceover

    Python 314 77

  4. manim-euclid-elements manim-euclid-elements Public

    A video rendering of Euclid's Elements using Manim

    Python 15 4

  5. janitr janitr Public

    Local runnable models for filtering out crypto shill content, AI replies, and other undesirable content

    JavaScript 69 2

  6. ghzinga ghzinga Public

    Easy, clickable TUI to view a single GitHub issue or PR, in Rust

    Rust 82 3