Havuz

AI native API/MCP framework for the agent era. Built with agents, for agents.

CLAUDE.md + AGENTS.md 12-rule behavior file for Claude Code, Codex, Cursor, Hermes — drop-in, ~700 tokens, cuts AI coding mistakes ~40%→~3%

A multi-tenant vector database focused on extreme RAM efficiency. Lightweight, scalable, and optimized for high-density deployments.

A Claude Code skill that rewrites your prompts for clarity before they run.

Project continuity for Claude Code and Codex: a roadmap beside your code, grounded handoffs, and clear task status.

Geode provides agents remote access to your Obsidian vault.

Multiplayer infrastructure for AI agents

Audit, clean, and harden a WordPress site end to end, with an AI agent (Claude/Cursor/Codex/MCP or SSH). Detects & removes hidden malware —

Argus is a desktop app that helps find and fix wasted agent work. Local, free and open source. Focuses on business tasks, not code. Works wi

Typed structured extraction from PDFs, photos & text via any LLM — Instructor for Swift, document-first

Macaw — on-device AI agent for macOS. 2.7B LFM2.5 fine-tune, MLX 4-bit, 97 macOS tools, no cloud.

A skill library that teaches Claude Opus the working disciplines of Claude Fable 5. Move the checkpoints, not the capacity.

Lightweight, dependency-free logging and analytics for LLM API calls — track tokens, cost, and latency to a plain JSONL file.

A runnable reference implementation of multi-agent constraint propagation using Caura. 5 specialists share memory via MCP so decisions made

A customizable Claude Code setup for knowledge workers: agents, safety hooks, skills, memory, and an Obsidian knowledge base. Clone, run set

Your AI finally remembers your codebase

Run Claude Code on any LLM provider without losing your setup, local-first proxy with verified profiles, OAuth device-flow login, failover,

Live DevTools for your running React + TanStack app — driven from the terminal by your AI agent, through one CLI

Transparent, transport-layer semantic cache for LLM API calls, powered by Redis 8 Vector Sets.

End-to-end educational RAG system: dual-engine retrieval (BM25 + BGE-M3 hybrid vector search), adaptive query strategies (HyDE/sub-query/bac

See your AI coding costs in real time. Tokenline adds context usage, prompt-cache savings, TTL countdown and rate-limit pacing to Claude Cod

Ambient macOS light strip for AI coding agents — a Dock-adjacent pill of eight LEDs showing whether anything is working, waiting, finished,

Skills that get AI assistants to stop agreeing with bad product ideas by default — pressure-test premises, separate hobbies from businesses,

The only SEO audit CLI that reads your codebase first, then pulls real Google Search Console data. Free, open source, bring your own LLM.

Production Claude Code skills for n8n from a Verified Creator's 100+ workflows

LLM-powered conversation summarisation service for OCTO — turns group chats and threads into structured briefs with key decisions, open ques

khive is a system of records and operations for agent-native organizations

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

Browser extension with 35+ rules for keeping your AI agent safe while browsing

An Agent with Semantic Memory. This is an experiment for Neuro-Symbolic AI

A minimal, evidence-gated harness for Codex and Claude Code: reuse what works, add only what is missing, verify before evolving.

Personal fork of czlonkowski/simple-meeting-scribe (MIT) — local macOS meeting transcriber: Ukrainian support, summary export (Telegram/Obsi

Provider-neutral bounded agent runtime for Android/JVM: controlled context (trust/priority/budgets), profile-scoped tools, OpenAI-compatible

Turn slide screenshots and generated images into editable PowerPoint decks with visual-layer splitting, OCR evidence, and QA.

Privacy-first bilingual webpage translation browser extension.

为 Token 燃烧而生：token 越来越便宜，质量永远稀缺——与其省 token，不如把它当燃料烧。同一道题派最多 1000 个 agent 并行竞赛、裁判择优，用冗余换质量。Go 写的开源终端 Coding Agent，类 Claude Code，可接入任意模型，对团队友好

Use third-party OpenAI-compatible models with VS Code Copilot Chat. Includes proper reasoning support, usage tracking, etc.

Chain two LLMs — local or cloud: a Prompter that refines your idea, and a Coder that writes the code.

The gate between an AI agent and real money. Runtime authorization for financial AI agents: structured intent, deterministic policy, human a

The infrastructure for the AI economy: primitives for inference, agents, and decisions.

Token-efficiency analyzer for AI coding agents (Claude Code, OpenAI Codex CLI): cost estimates, cache analysis, and savings advice. Zero dep

Coding Agent insights for teams

Make any Claude model adopt Claude Fable 5's working style in Claude Code — always-on output style + zero-dependency MCP + subagent injectio

AI agents now operate with authority. Authority without discipline is how complex systems fail. Nuclear’s control loop, ported to AI-assiste

Local-first AI workflow orchestration for chaining models, agents, tools, and scripts into repeatable workflows.

Skills for your skills. Create, test, audit, package, and visualize Agent Skill libraries—locally.

Itemized cost receipts for Claude Code and Codex — by turn, verified, local.

AI-powered personalized IELTS corpus distillation — install-only skill (zero-backend Agent loop) + runnable backend with QMD-RAG. 7-stage pi

Use ChatGPT/Codex OAuth in Claude Code without an API key