潜力新星（90天）

Evolutionary multi-agent runtime that breeds, evaluates, and improves autonomous agents across reproducible epochs to converge on optimization of a goal.

macOS desktop widget for OpenAI Codex usage, quota tracking, token usage, and today task board

audio.cpp with a full-task WebUI - pure C++ audio-model inference engine powered by ggml. TTS, ASR/STT, VAD, voice conversion, speaker diarization, music generation. No Python dependency.

a community oriented 1:1, vLLM-alike (Continuous batching, paged KV) engine in C++ with additional features (GGUF, RadixAttention, Cache-aware scheduling, ...)

Open-source CLI, API, web UI, and reproducible benchmarks for probabilistic AI-generated image detection.

Video understanding and self-verification for AI agents. Turn videos, streams, and agent screen recordings into searchable, timestamped evidence—then use THE LOOP to inspect, fix, and verify the work. MCP, CLI, REST, local-first.

4DAnyone: Create Anyone in 4D from a Casual Monocular Video

An intentionally vulnerable OWASP LLM Top 10 training platform for AI Security, Prompt Injection, RAG Security, Agent Security, and GenAI penetration testing.

Local-first persistent memory for AI coding agents (Claude Code, Cursor, Codex) over MCP. Decisions, lessons and facts live in one SQLite file on your disk. Offline, multilingual.

HERO = Hashing · Edge cases · Rubrics · Overbuild — the four shapes coding agents over-defend in. A paste-in contract that stops them. Works with Claude Code, Codex, Antigravity, Cursor, Copilot, Windsurf, Gemini CLI.

LLMInternSkill: LLM internship resume and job-search Codex Skill for resume polish, JD tailoring, evidence guard, interview grilling, and Project Scout. 大模型实习简历与求职工具箱。

Agent Skill for turning Chinese classical poems into vertical Chinese-art videos with ImageGen stills, Docker I2V, calligraphy captions, retained ambience, BGM and final MP4 QA.

Refine and improve the quality of small faces in MiniMax H3 video. Per-frame face tracking, crop, refine with H3, stitch back.

Last-night exam-cram coach as a Claude Agent Skill: turns your slides, notes and past papers into a chaptered knowledge base + quiz bank, teaches only what's in your materials, and never fabricates (measured 100% out-of-scope abstention). Bilingual EN/中文 — the 期末极速备考 skill.

The cleanest, responsive AI workspace. Run highly tuned image and video workflows flawlessly from your desktop or your phone. Built on ComfyUI.

Extract Bilibili videos into learning-oriented Markdown notes with full subtitle and comment archives

AI Agent Skills for Chinese Knowledge Workers: iMandalArt, FIRE, planning, and publishing workflows for Claude Code, Codex, and LLM agents.

Local-first speech-to-text: no cloud, no telemetry, fail-closed by design. One CLI, seven model families, signed model catalog, OpenAI-compatible local API.

UmaDev: A coding agent that works like a real dev team, commanding the Claude Code / Codex / OpenCode you already use.

Honey (I Shrunk the AI) by GreenPT: a cross-tool coding skill that cuts AI coding-agent token usage and LLM API costs — write less code, less prose, and denser agent-to-agent handoffs (−53%, lossless in benchmarks) with no loss of quality. Works with Claude Code, Cursor, GitHub Copilot, Codex, Gemini CLI, Windsurf, Cline & Kiro.