Home
Curated open-source AI projects with verified GitHub data and hands-on breakdowns.
A 14,204-star Claude Code skill where a single /watch command downloads a video, extracts frames, transcribes it, and hands everything to Claude, so the AI finally sees the picture instead of only reading captions. #20 on GitHub weekly trending, MIT, Python.
earendil-works/pi (Pi Agent Harness) is a TypeScript AI agent toolkit split into five standalone packages - pi-ai, pi-agent-core, pi-coding-agent, pi-tui, pi-telemetry - with 84,648 stars and #10 on this week's GitHub trending chart.
HuggingFace's open-source voice agent pipeline with four swappable stages (VAD->STT->LLM->TTS), OpenAI Realtime-compatible protocol, can run fully local and fully open-source. #15 on GitHub weekly trending, 11,350 stars, Apache-2.0.
bojieli/ai-agent-book (33,213 stars, Python, Apache-2.0, created 2025-09-09, pushing today, GitHub Trending) is the open-source repo for the in-depth AI Agent book by Li Bojie: Design Principles and Engineering Practice. Agent = LLM + context + tools; 10 chapters from principles to engineering plus 95 hands-on experiments (local + external repro) plus 13 languages; PDF/EPUB free. The top pick for developers who want to learn AI Agent systematically; the moat is systematic depth plus experiments plus multilingual plus open and free.
nexu-io/open-design (84,003 stars, TypeScript, Apache-2.0, created 2026-04-28, still pushing 8/6) is the open-source Claude Design alternative and the agent-era Figma alternative. A local-first desktop app that open-sources the Claude Design agent-native loop (discover, lock, stream, critique, deliver) into skills plus DESIGN.md plus plugins, letting 25+ coding agents (Claude Code, Codex, Cursor) act as the design engine and produce brand-contracted single-page runnable artifacts (prototypes, decks, HyperFrames, images) exportable to HTML, PDF, PPTX, MP4. Latest v0.18.0 (8/5).
firecrawl/firecrawl (161K stars, TypeScript, AGPL-3.0, created 2024-04-15, still pushing 8/5) is an open-source web context API: a search/scrape/interact toolkit that turns any page into clean Markdown or structured JSON for LLMs and agents. Vendor claims 96% web coverage, P95 3.4s; one command plugs into Claude Code/MCP. Cloud starts ~$16-19/mo; self-host is free but some features are cloud-only.
video-shotcraft (Vincentwei1021/video-shotcraft, 3.5K GitHub stars, TypeScript, Apache-2.0) is an agent skill that turns Claude Code or Codex into a motion-design studio - point it at your product and it storyboards, animates, and sound-designs a cinematic promo with Remotion. Ships 104 shot recipe cards, 161 motion previews, a validated 36.2s Ink Press template, 2.5D camera moves, beat-synced cuts, and film-grade SFX. Output is reproducible TSX, not black-box generation. Includes a China AtomGit mirror and 3 headless-rendering CI flags.
book-to-skill (virgiliojr94/book-to-skill, 16.2K GitHub stars, Python, MIT) distills technical books/PDFs/EPUBs/doc folders into structured agent skills following the open Agent Skills standard - install once, works across GitHub Copilot CLI, Amp, and Claude Code. Generates SKILL.md + per-chapter files + glossary + patterns + cheatsheet, with chapters loaded on-demand so they don't count against your token budget. Ships a benchmark tool measuring 24-51x fewer tokens than dumping the full book into context (tested on 3 real books). Beyond books: internal docs, brand systems, research clusters, specs. Includes a copyright-compliance note.
alibaba/open-code-review (18,287 stars, 1,230 forks, Go, Apache-2.0, created 2026-05-18, pushed today) is Alibaba's open-source hybrid-architecture AI code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE/thread-safety/XSS/SQL injection), OpenAI/Anthropic compatible, originating from two years of large-scale internal validation at Alibaba, with Claude Code/Codex/Cursor integration and CI/CD support.
h4ckf0r0day/obscura (19,954 stars, Rust, Apache-2.0, created 2026-04-13) is an open-source headless browser engine for AI agents and web scrapers, written in Rust with embedded V8, 30MB memory, 85ms page load, built-in anti-fingerprint stealth mode with 3,520-domain tracker blocking, CDP protocol for Puppeteer/Playwright compatibility, and a built-in MCP server for direct AI agent integration.