추천 AI·머신러닝 오픈소스 프로젝트

An open source AI coding agent that runs in your terminal, with a desktop app and two built-in agents for building and planning changes.

A Python framework for recursive harness optimization that refines the prompts, tools and orchestration around a fixed LLM agent on unseen tasks.

A Claude Code skill that digs into unfamiliar topics citation by citation until it reaches primary sources, saving them unsummarized with claim-level citations.

Lets an AI vision model operate a real iPhone using a $7 ESP32-S3 board that acts as an AirPlay screen receiver and USB HID mouse, with no computer needed at run time.

A browser-based AI image upscaler that runs Real-ESRGAN and Swin2SR models locally, so images are never uploaded.

A reinforcement-learning GuanDan (Chinese card game) AI with PPO training, plus an open-source browser game service to play against it.

A Claude Code skill that reads your local session logs to explain where your token usage went and why you hit a limit.

An open-source AI productivity workspace built on Cloudflare Workers, where agents build and share small custom apps called Gadgets under company-specific guardrails.

A free, two-way Figma MCP server that lets coding agents read Figma designs as code and write designs back to the canvas via a plugin.

Self-hosted auth gateway that lets AI agents call 1,000+ SaaS providers through SDK, CLI, MCP, HTTP, and OpenAPI without handing over credentials.

Open-source database and workspace where AI agents and people share structured records, docs, skills, and apps, with reviewed writes.

Open-source platform combining AI agents and workflow automation with 250+ connectors, self-hostable or embeddable in your SaaS.

Open-source Python library for building AI agents that control a real browser, with optional Browser Use Cloud stealth browsers and hosted agent API.

Open-source AI gateway giving one OpenAI-compatible API across 23+ LLM providers, with failover, load balancing, semantic caching and virtual keys.

Open-source Go/Python runtime for serverless AI workloads, offering GPU inference endpoints, sandboxes and background task queues with scale-to-zero.

AutoGPT is an open-source platform for building, running, and scheduling AI agents that carry out multi-step workflows from plain-English descriptions or a visual builder.

Open-source, self-hosted AI observability platform for tracing, evaluating, and debugging LLM applications and agents.

All-in-one AI app for chatting with your documents and running AI agents, with local-first setup, multi-user support and many LLM providers.

A Python framework and runtime for building, serving, and managing AI agent platforms, with a web UI for administration.

An open agent framework where the AI agent gets its own Dockerized Linux desktop, browser, and document tools to carry out real work.