Havuz

A dependency-free Python genetic-algorithm engine with pluggable fitness criteria — the reusable search core for an LLM/AI autoresearcher.

Bounded, inspectable LLM inference pipelines from declared YAML — runs offline against Ollama or any OpenAI-compatible local server, emittin

CoDA-Bench is a benchmark for code agents on data-intensive tasks. 🎈代码智能体能搞定数据密集型任务吗?

From Pipeline to Council — A lightweight agentic framework for search & recommendation. 五个 Agent 重构搜广推管线。

Local-first AI research assistant: plans sub-queries, searches the web, and writes fully cited reports. LangChain + LangGraph multi-agent +

Validate, repair, and retry LLM structured outputs. 13 repair strategies for common JSON malformations, JSON Schema validation, and retry-wi

An LLM-powered social simulation and roleplaying engine where NPCs have their own lives too.

Task-plan driven SEO and LLM-friendly website skill cluster

Private, self-hosted document chat for attorneys: parse legal PDFs and query them with local open-source LLMs (Ollama) + verifiable page cit

OpenAI API-compatible LLM Model powered by YOU

Mindlas catches your coding agent drifting before the bad code lands. Real-time tracking and correction for context rot, unverified done cla

Wavelet-based multi-resolution context management for LLMs via MCP

把一批文章样本自动炼化成可复用的写作风格档案，并生成一个按该风格写作的专属 Claude skill。/ A Claude Code skill that distills writing samples into a reusable style profile and for

Write commit messages, grep by meaning, turn English into shell — small composable CLI tools backed by an LLM. Local by default (no API key)

Research-first architecture engine for Python, TS, Go and Rust. Mines GitHub Issues for real production failures before scaffolding, then au

Give Claude Code a persistent memory — it remembers you, your projects, and your decisions as plain Markdown files you own. Fully local. Jus

Self-hosted meeting transcription portal — speech-to-text, speaker diarization, LLM-corrected transcripts, structured summaries and Word min

Free unlimited Gemini API for any OpenAI-compatible app.

A unified OpenAI / Anthropic / Responses API gateway with built-in vision model injection — let any text-only model "see" images.

RAMen is a fast in-memory data store like Redis, but built for AI: drop-in Redis protocol, native vector search, semantic caching, and a bui

Protocol-layer harness for DeepSeek: Python witness stack — posterior verification that keeps the protocol honest. dsh doctor --node probes

Dark, lime-accented one-page portfolio of Bharani Kumar S — an AI full-stack developer. Smooth-scrolling vanilla JS, SEO-ready, Resend conta

One command to cut token usage by up to 50%+

RepoPilot AI is a single-page web application that uses AI to analyze GitHub repositories and provide instant insights into architecture, te

Your AI works the night shift — read-only reviews of your projects while you're busy, one digest every morning.

Extends pageIndex into an AI document workspace with multi-format parsing, OCR, visual TOC, custom models, citations, and agentic QA.

Reina 桌面 AI Agent 发行仓库（仅安装包） | Release distribution for the Reina desktop AI agent — installers only. Windows x64 & macOS arm64, in-app auto

Your agent team, on your machine. Open-source harness for multi-user Claude agents.

Shorter replies with Claude Code. Save output token spend.

Run AI models on Apple Silicon and Raspberry Pi

Mobile-native agent SDK for building on-device AI agents with sandboxed execution, extensible tools, and x-app / x-agent / x-channel connect

Cross-architecture LLM internal observation database (23 models, 13 architecture families). Exposed as MCP tools for any AI coding agent.

CVE ingestion + enrichment pipeline with an MCP server for AI agents.

OpenCode agent configuration — a read-only orchestrator delegates to 6 specialized subagents. Expert model plans, fast models execute. Isola

Translate and rewrite without switching apps. Free, open-source, on-device writing for Mac. No account or API key.

A discipline plugin for Claude Code + OpenAI Codex — plan first, verify against the real code, receipts on every decision, independent adver

CPU-side action routing, context compression, and causal memory for AI agents — matches an LLM-everything agent at 58% fewer LLM calls and ~

Local-first CLI for benchmarking LLMs on real hardware — quality, speed, reliability, and a real multi-turn agent loop.

Find and fix web-app vulnerabilities in one command — SAST + DAST + AI, for devs who aren't security experts.

Session memory for Claude Code. Auto-captures every session into a plain Obsidian markdown vault. Zero config. MIT. macOS + Linux.

Eve Agent V2 Unleashed — local-first autonomous AI coding agent powered by Ollama

"The First Principles of AI Programming" Open Source Book. Starting from the underlying physical constraints of large models, it deduces the

Every way an LLM can break one SDK to stop it. Injection. Hallucination. Leakage. Handled.

SGLang Docker image for AMD Strix Halo (gfx1151 / Radeon 8060S)

Papers in, ideas out. Generate research, product, or feature ideas from 250M+ academic papers.

Track brand citations across seven AI search surfaces. Free OSS/self-host; optional EUR 99 client-ready audit.

LLM-powered Chrome extension for automatic web form filling. All data stays local.

From Agent to Agency — 一个 AI 做不了的事，一群 AI 可以。Multi-Agent Collaboration Platform.

Deterministic agentic-style multi-hop RAG at research-SOTA parity on commodity LLM APIs — no GPU, proof tree per answer.

Translate books and documents with LLMs, featuring glossary extraction, review, and polishing pipelines.