Havuz

Contacts for Claws

Simple and composable logging for Swift apps, servers, and coding agents.

Apple Silicon mlx with Zero Dependency Java

基于pi agent core构建的harness agent

Open-source always on personal agent with its own computer. It remembers your work and keeps going while you're away. Works with any model.

Type-aware, evidence-backed delivery contract for AI coding harnesses. Different agents in. One predictable human handoff out.

Prove your agent skills work before you ship them.

Kyros — The Memory OS for AI Agents Give your AI agents secure, self-correcting, persistent memory in 3 lines of code. Three memory types

Turn Claude Code into a senior software architect that orchestrates a team of specialist agents and audits itself. An 8-phase skill with TDD

Official Repo for Spreadsheet-RL

ZeroSpec is a zero-dependency Markdown framework that helps AI agents understand your repository structure, rules, and source of truth befor

Agent Skill that turns your coding agent into a keyboard firmware expert. Audits ZMK/QMK keymaps, tunes home row mods, makes trackballs laye

Fast structured Android control loops with TypeSafe Jev and Mobile MCP

An intelligence layer grounding autonomous agents in verified, real-time knowledge at scale.

SkillRoute by JEStats — local-first skill routing for coding agents, with ranked matches, confidence, and evidence over MCP.

Tool-to-Agent Protocol: tools can be smart without embedded LLM calls.

Agent Skills official site

Drop-in replacement for PostgreSQL with surgical, dependency-aware undo of transactions. Makes incident recovery zero-cortisol.

Introducing Muse Glimmer: open-weight 30B agentic multimodal model that runs on your device (Meta). Interactive local agent lab + guide. Apa

Keep growing TypeScript codebases maintainable. Trellis measures complexity, duplication, and import cycles, pinpoints structural debt, and

Own your LLM's web search: a local search->fetch->rank pipeline that replaces hosted web-search tools. Measured: matches hosted accuracy at

Control plane for swarms of sandboxed AI agents: run, fork, pause, steer, and checkpoint agents that each work inside their own isolated san

Local, self-hosted agent infrastructure — one endpoint for every AI client (Claude Code, Codex CLI, Gemini CLI, OpenCode). Unified model acc

An LLM-friendly, single-binary, drop-in replacement to Supabase

Iterative agent harness improvement: run a coding agent on a hard task, generate the reusable tooling it was missing, qualify it, and replay

Decision harness for TypeSafe Jev — confidence gates, shadow mode, recipes, and evals. Claude CLI 48.9s → Jev 1.3s on the same row-filter jo

多agents协作平台

A locally-running multi-agent human in the loop system for deep, interdisciplinary research with only 4 dependencies.

Distributed, objective-led control plane for Lean 4 formalization projects.

Open-Ended Long-Horizon Text Game Generation for Test-Time Continual Learning Agents

Multi-agent coding factory made fast.

Self-hosted 多用户 WeChat / 微信 AI 智能体控制台，支持 Web 管理、Docker部署、MCP、Skills、记忆、定时任务、沙盒执行和长期运行，支持分享AI Bot给团队成员。

PatientAgentBench: A Benchmark Framework for Evaluating Patient-Facing Health AI Agents

TypeScript AI agent runtime. Write your agents in code and choose how predictable they are: deterministic steps where they can be, an LLM on

Vetix — Automated scanning, identification, and assessment of SKILL security risks, based on Golang.

A cross-platform agent framework and samples, support on Windows/MacOS/Android/iOS/HarmonyOS

a unix-like du command line tool to count token usage per files and directories

Geode provides agents remote access to your Obsidian vault.

Internal agent platform on Amazon Bedrock AgentCore: interactive Claude Code cloud workspaces + headless agent kernels behind one portal, wi

AI-native, source-available (PolyForm Perimeter) game-dev toolkit: bforge — deterministic headless-Blender asset forge for AI agents (138 op

Agent Usage Detail Record — an open JSON standard for attributing and costing agent runs

AiGuard —— 本地优先的 AI 流量安全网关：本机 MITM 代理只接管 AI 域名，请求侧脱敏、响应侧流式还原，并识别七类风险信号。原文不落盘、不上传。（注意如果你需要模型直接使用token或者其他密钥进行操作，请放入白名单或关闭该功能！！！！）

长任务执行 Skill：先定成功标准再逐轮验收，让 Claude Code/Codex 跑完长任务不半途而废｜Goal-driven long-running agent tasks

Secure and scalable way to delegate work to AI agents from anywhere. Build a team of agents that work for you.

🛠️ toolhub - Turn ANY console script into an LLM tool in 2 seconds. Lightweight, self-hosted, tree-structured alternative to MCP & LangChain

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

Is your CLI ready for AI agents? Automated checks to ensure optimum agentic CLI use: don't block agents, don't waste tokens, ensure safe use

Mission control for every Claude Code session on your machine.

A minimal, explicit, provider-safe harness for the Pi coding agent