#typescript

OMX is a workflow layer that enhances OpenAI Codex by adding better task routing, reusable skills, and project state management. It keeps Codex as the execution engine while introducing canonical workflows like `$deep-interview` for clarification, `$ralplan` for planning approval, and `$ralph` or `$team` for execution. The benefit is that you get a structured, consistent approach to complex coding tasks—clarify intent first, approve the plan, then execute with either persistent single-owner loops or coordinated parallel work—without replacing Codex itself, just making your day-to-day work more organized and efficient.

https://github.com/Yeachan-Heo/oh-my-codex

#typescript #electron #open_source #pixijs #screen_capture #screen_recorder

OpenScreen is a free, open-source app like a simple Screen Studio for making beautiful product demos and screen walkthroughs. Record your full screen or windows, add zooms, mic/system audio, backgrounds, motion blur, annotations, trims, speed changes, and export in any ratio. Download from GitHub for macOS, Linux, or Windows—it's MIT-licensed so you can use, modify, or sell it freely with just a copyright notice. You save $29/month, get core tools without limits for personal/commercial use, and customize fully.

https://github.com/siddharthvaddem/openscreen

#typescript

QMD is an on-device search engine that indexes your markdown notes, meeting transcripts, docs, and knowledge bases for fast keyword, semantic, or hybrid searches using local AI models—no internet needed. Install via npm or bun, add collections like `qmd collection add ~/notes --name notes`, embed with `qmd embed`, then query like `qmd query "project timeline"` for top results with scores and context. It integrates with AI agents via JSON output or MCP server. You benefit by quickly finding info across your files to boost productivity and make smarter decisions in agentic workflows.

https://github.com/tobi/qmd

#typescript

Multica is an open-source platform that turns coding agents into real teammates. Assign tasks to them like colleagues—they write code, report issues, update progress, and build reusable skills over time, with no babysitting needed. Use Multica Cloud for instant start or self-host with Docker and CLI for local control; it works with Claude Code, Codex, and more. This saves you hiring costs for your next 10 developers, boosts team productivity, and lets humans and AI collaborate seamlessly on the same board.

https://github.com/multica-ai/multica

#typescript

Ralph is an autonomous AI agent that loops coding tools like Amp or Claude Code to fully implement your project's Product Requirements Document (PRD) by tackling one small user story per fresh iteration, using git history, progress.txt, and prd.json for memory. Setup is simple: install prerequisites, copy scripts or skills to your repo, generate a PRD, convert to JSON, then run `./scripts/ralph/ralph.sh` for up to 10 iterations until all tasks pass checks and complete. This saves you hours of manual coding on greenfield features, delivering working code reliably with minimal supervision.

https://github.com/snarktank/ralph

#typescript #ai #cuda #mlx #qwen3_tts #qwen3_tts_ui #voice_ai #voice_clone #whisper

Voicebox is a free, open-source voice synthesis studio that lets you clone voices, generate speech in 23 languages, and apply audio effects—all running privately on your computer. You can create realistic voice clones from just seconds of audio, use five different text-to-speech engines for different needs, add effects like reverb and pitch shift, and build multi-voice projects with a timeline editor. The key benefit is complete privacy: your voice data and AI models never leave your machine, unlike cloud-based alternatives. It also includes an API for building voice-powered applications and works across Mac, Windows, and Linux with GPU acceleration support.

https://github.com/jamiepine/voicebox

#typescript #agent #agents #ai #background_agents

Open Agents is an open-source tool that lets you build AI coding agents running on Vercel. It separates the agent logic from the sandbox environment, meaning the AI runs outside the virtual machine and controls it through commands like file edits and shell operations. This architecture gives you flexibility—the agent isn't tied to a single request, sandboxes can pause and resume independently, and you can swap AI models or sandbox implementations without rebuilding everything. You get chat-driven coding with file tools, durable multi-step execution, isolated sandboxes, GitHub integration for cloning and creating pull requests, and optional voice input. The benefit is a scalable, modular system you can fork and customize for your own background coding agent needs without managing your laptop.

https://github.com/vercel-labs/open-agents

#typescript

Craft Agents is a free, open-source desktop app for easy AI agent work with Claude and others. Install in one line, connect APIs like Slack or Gmail by just telling it—no setup hassle. It supports multitasking, sharing sessions, custom skills, and automations in a smooth UI. Run headless on servers too. You benefit by saving hours on configs, boosting productivity with intuitive tools for powerful AI tasks anytime.

https://github.com/lukilabs/craft-agents-oss

#typescript

T3 Code is a simple web GUI for coding agents like Codex and Claude. Run it instantly with `npx t3`, or install the desktop app via winget (Windows), brew (macOS), or yay (Arch Linux)—but first authenticate a provider like Codex CLI or Claude. It's early with bugs, no contributions yet; join Discord for support. You benefit by quickly coding with AI agents in a minimal interface, skipping complex setups for faster, easier development.

https://github.com/pingdotgg/t3code

#typescript #ai #ai_agents #llms #on_device_ai

Thunderbolt is an open-source AI client you control fully—you pick models like Ollama or llama.cpp, own your data, and avoid vendor lock-in. It runs on web, iOS, Android, Mac, Linux, and Windows, with enterprise features for on-prem deployment. Still early (needs your model providers and optional backend), but self-host via Docker to test. This benefits you by giving private, flexible AI access anywhere without big-tech dependency, cutting costs and risks.

https://github.com/thunderbird/thunderbolt