Havuz

Benchmark llama.cpp on your own GPU/CPU boxes and compare against everyone else's — sweep runner, an sha256-keyed index of ~4M GGUF files, a

A Go implementation of Claude Code — This project was built entirely by a team of Claude AI agents working in parallel

An AI chat agent (harness) with integrated tools, skills and an optional knowledge base. Chat with any OpenAI-compatible LLM, use built-in i

Per-step LLM routing benchmark with 970 static labels, live SWE-bench evaluation, an open data pipeline, and a public leaderboard.

A personal substrate for agent packages that ship their own screens. One manifest declares the agent, its verbs, and its UI.

Turns an organization's internal docs into a living learning system: AI agents generate courses, tutor learners, and adapt content on the fl

RepoDoc is a lightweight, zero-dependency utility that scans any project directory and generates a complete Markdown document containing the

Research-backed Harness Engineering study repo with a runnable Python MVP built around Plan -> Work -> Review.

Document extraction and validation for logistics companies.

Open-source control plane for private, governed enterprise AI — one gateway for every model, composable governed pipelines, and apps your wh

Self-hostable agentic OS. Build, run & audit multi-agent AI teams from signed plugins. Bring your own LLM key, own all your data, EU/GDPR-re

Guard rails for Claude Code: workspace boundaries, protected branches, production targets, lost exit status, and foreground time.

Private AI WYSIWYG Markdown editor with agentic chat & voice dictation. Open source, no cloud, no tracking.

Scaffolding skill + always-on conventions (CLAUDE.md/AGENTS.md) for Quarkus + LangChain4j agentic AI apps: AI services, multi-agent workflow

A small operating file for keeping long-running AI agent sessions from degrading. Capable models can do useful long work. The problem is no

Local-first YouTube research: acquire VTT, search with SQLite/FTS5, run human-gated cumulative research in a web app or CLI, query via read-

Simplified Technical English for Code - extract, adapt, and enforce ASD-STE100 rules for software documentation. 9-agent pipeline, 59-test b

CLI to generate a narrative walkthrough of any diff

GitHub discovery & LLM evaluation engine with automated follow, star, and auto-unfollow management.

Single-HTML monitoring dashboard for llama.cpp — per-slot sampling, plain-language tuning hints, router-mode aware, with built-in chat. Zero

Proxy MCP: one server that runs Playwright, Figma, Chrome DevTools, Shopify Dev and Context7 behind a single connection - and cuts their tok

Detect hallucinations in LLM responses. Verify every claim against source documents using hybrid STS + NLI. Works with LangChain, LlamaIndex

Cross-vendor agent-to-agent protocol — Claude, Codex, and Gemini communicate via file-based P2P messaging.

An Always-On Autonomous AI Software Engineering Team built with LangGraph, FastAPI, and Next.js that operates natively within your GitHub re

デバッグ能力を「鍛える」と「測る」を1つにしたBtoB向けの仕組み。AIは答えではなく段階的な問いを返し、業務を止めずに技術力が育つ。到達度・正答率・成長率を5軸で可視化し、人事評価の客観的な根拠として使える。

Self-hosted MCP server + dashboard giving Claude Code, Cursor & Hermes agents shared long-term memory over a git-backed markdown vault. Agen

Open-source realtime voice AI with adaptive VAD, latency telemetry, interruptible TTS, an inspectable TypeScript WebSocket gateway, and expe

Use third-party and Chinese model APIs as bounded Codex subagents while keeping Codex as the main agent.

Chat with your OpenClaw agents from Open WebUI and watch them work: tool calls as cards, questions as dialogs, sub-agents in a live panel.

Your agent can read git log. It can't read the shell commands you tried that failed. NexusMem indexes shell history with exit codes, git com

Adversarial multi-turn benchmark for LLM debate quality, using side-swapped matchups and multi-model judging to rank models by judged debate

contextzip CLI: zip the right parts of your codebase for ChatGPT, Claude or Gemini, then apply the AI’s changes back. Skips dependencies, re

Zero-config local LLM optimization for Ollama, LM Studio, and Apple Silicon MLX. Reduces TTFT by 40%, wall time for local agents by 46%, and

Bittensor Subnet Repository maintainer agent + a GitHub-history benchmark that scores it — can an agent make the maintainer decisions a stro

12 agent skills for AI-driven software development — a gated idea-to-ship pipeline: 7w3 design, build-ready specs, executable architecture,

Lightweight AI agent runtime for homelabs, built around deterministic skills and local LLMs.

A 125-task benchmark for knowledge-mediated implicit associations and retrieval blind spots in long-term agent memory.

Free, open-source AI coding agent & multi-agent teams for macOS — runs 100% on-device via Ollama or LM Studio. Autovisor, an autonomous Supe

Self-hosted chat client for Ollama, OpenAI, and Anthropic. Web UI and terminal CLI, with web search and a browser-automation agent.

Runtime economic governance for AI agents. Circuit Breaker adds risk-managed execution, cost ceilings, and kill switches to autonomous AI wo

The open-source, local-first alternative to LangSmith. No cloud. No setup.

High-throughput, self-hostable LLM router in Go. BYO provider keys, k8s-native, key pooling + batch + usage observability.

inplan — a Markdown editor for human ⇄ coding-agent planning. CLI + desktop editor.

Manages parallel Claude Code sessions, so you don't have to.

Continuously refine your project’s engineering philosophy — and keep AI coding agents honest with evidence-backed completion.

Local monitor for Claude Code API traffic, logs, prompts, and tools

Six steps to make a new Phoenix app work well with AI coding agents: pin the toolchain, turn mix precommit into a real gate, split the gener

Turn documents into high-quality instruction datasets with grounding, quality filtering, deduplication, and provenance.

Native iOS app for talking to your OpenClaw agents by voice or text. On-device speech recognition, streaming responses, multi-agent channels

A local-first coding agent runtime, multi-provider LLM gateway, and terminal workspace — built from scratch in Go.