潜力新星（90天）

个人 IP 图解创作器：用你的 IP 形象，生成专属内容图解。

Pharos — local-first agentic RAG for your team's document library: multi-format ingest, hybrid retrieval, enterprise ACL, dual HTTP + MCP exits.

Matt Pocock 技能集的中文翻译版 — 地道中文，原汁原味的技术术语。基于 mattpocock/skills 复刻。每日中午12点钟同步

Custom APRL machine-learning algorithm + a curated ML knowledge base of 923 papers, lectures, and explainers. NumPy classifier/regressor, tests, benchmarks, provenance, Obsidian, and agent-ready Markdown.

Opus에 근접한 품질을 Sonnet 단가에 — 가성비 AI 운영 모델을 직접 구축한 Claude Code 프로젝트

To know what models don't say out loud.

AI character studio on a node canvas. Build consistent characters, train LoRAs, and generate on your own GPU.

🧠 The Brain for Your AI — Local-first memory engine for AI agents. Store, recall, and search memories with semantic embeddings. Single Rust binary, zero config, fully offline.

macOS AI 语音输入工具，支持中文与中英混合识别、AI 文本整理、上下文格式匹配和本地优先。

Jacobian-Brainwash : A manual alignment tool for large language models built on Anthropic's Jacobian Lens. Results are exportable.

Boot your PC straight into an LLM. Rust, UEFI-resident, no operating system underneath.

Universal, model-agnostic operating harness for AI agents (Claude, Codex, Gemini, …) — a lean core + work-type profiles assembled by one setup script.

hh.ru аи бот для откликов

Mila — native macOS local transcription app (whisper.cpp) with optional speaker diarization. Apache-2.0.

Provider-pluggable orchestration runtime for multi-model AI inference. ( Sakana Fugu style )

Local-first Memory OS for personal AI assistants with L0-L3 memory, Wiki++ knowledge, skill routing, and TokenLess context compression.

Local proxy that compresses your LLM API requests so you pay less, with no change to the answers. Trims wasted tokens from prompts, history, tool output, and code before they're sent: -31% input / -74% output, measured live. Any provider, no extra model calls. Also an MCP server and embeddable library (Rust, Python, Ruby, Kotlin, Swift, JS/TS).

Self-hosted, one-tab workbench for the whole LoRA lifecycle: build Character / Concept / Style datasets (generate from a reference, scrape, or triage a huge dump in the Image Bank), curate, caption, clean watermarks, train locally or in the cloud across five model families, then rank checkpoints in a Test Studio. Free & local.

Open Dungeon — the first easy-to-use, fully local AI roleplay app. Story and inline scene images generated 100% on your machine (Gemma 4 QAT via Ollama(and others) + FLUX). No accounts, no API keys, no cloud.

🚀 50 production-ready Generative AI SaaS apps — brand them, ship them, keep 100% of the revenue. Stripe billing · Google OAuth · Vercel deploy · MIT licensed