Havuz

Personality layer for coding agents. Same code, different voice: yo / dawg / mafa / off.

Hermes Agent context engine: the main model writes the compaction handoff on its cached prompt prefix

A small, free, open-source Mac and Windows app for talking to your AI agents, WhatsApp-style: OpenClaw, Hermes, Claude Code, Codex and any A

A live 3D sky city of your Claude Code agents. Local-only.

EvoHarness-RL: Learning Runtime Harness Coordination for Self-Evolving Agents

A typed, permission-aware model of your business data.

Native iPhone companion for self-hosted Hermes, with a private authenticated mobile bridge.

Fast Rust launcher for DeepSeek Harness, with clear Claude Code/Codex migration, model setup and lightweight local tools

Weft Coordination Protocol (WCP): live coordination for parallel coding agents across harnesses, built on Cloudflare. An Atelier (elier.ai)

Build your own personal AI agent, one mission at a time. Mission 01: your AI learns who it's working for.

Open Instinct, iMessage ve e-posta üzerinden kullanılabilen, bilgisayar ortamında görev yürüten ve diğer kişisel agent’larla iletişim kurabilen bir kişisel agent altyapısı sunuyor.

Neden yükseliyor · Repo çok yeni olmasına rağmen 126 yıldız ve 28 gün içinde 27 commit ile dikkat çekiyor. Kişisel agent, gerçek görev yürütme ve agent’lar arası güvenli koordinasyon fikrini tek bir dağıtılabilir projede birleştirmesi ilgi çekici; ancak yükselişin nedeni ve gerçek kullanım düzeyi verilen ölçümlerden doğrulanamıyor.

Privacy first B2B AP agent that handles Business invoices, approvals and payments with controlled policies.

Run AI agents in a sandbox that restricts what they can execute, read, write, and reach on the network. Box combines OS isolation with defau

Official implementation of “ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation”

Connect AI agents to 1,400+ APIs across search, social, SEO, finance, and more. Automatic tool discovery, unified billing, zero subscription

PACT is an open protocol for personal agents to interact with businesses’ agents under explicit, verifiable user permissions.

Local decision model with calibrated probabilities: send a state and yes/no, choice or score questions, get a probability for every option.

A living AI civilisation you can watch: tiny creatures discover, build, teach and pass knowledge on, with any LLM in their minds.

⚡ Fastest and lightweight harness for AI agents

Open-source AI workspace for humans and agents — SuperCompress as the core context layer.

AI agents and workflows for Kotlin Multiplatform, as small typed graphs: loops, parallel steps, streaming, save points and human approval. I

Open decision models from 0.8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. Engine f

先进且稳定的企业级智能分析平台；面向企业问数、洞察分析的Data Agent，基于数据分析方法论打造可审计、可治理、可复现workflow

An always-on personal agent you own, on the model you choose. A model-agnostic alternative to OpenAI's Dots, built on OpenComputer Serverles

Voice-first desktop assistant for Windows — talk to your PC and it works inside real apps, safely. By VORNEX.

Your AI agents, always on, working together. Claude Code, Antigravity and Codex on your own computer.

Tiny pixel coworkers that live in Paseo and never ask for a raise. A Grok Bot / OpenMausBot alternative plugin for https://paseo.sh/

DevLabs™ Free, open source and local-first. A multi-agent harness that works with the AI subscriptions you already pay for. It turns the ter

A persistent workspace for development work.

Small enough to run on your Mac. Smart enough to ask. An open-source computer use agent for macOS, with local models (MLX) that see, decide

Agent Factory is an open-source studio for building, editing and chatting with AI agents. You build them from simple nodes: LLM calls, HTTP

A 4B decision model: calibrated probabilities for yes/no, choice and score questions over long inputs.

Stringhand gives AI agents a pair of hands for the guitar.

A production-grade agent runtime engineered from the Harness Engineering philosophy.

Help your agent remember what matters. Keep preferences, carry context across sessions, and bring your memory to the next agent.

Multi-platform IM channel bridge with unified message abstraction for AI agents and bots

Agent-first lightweight browser automation — direct CDP, no Playwright, more accurate and more reliable.

Working-draft Mail Action Protocol for agent actions through email, with type contracts, a registry, and a validation CLI.

A fast, lightweight CLI.

A host-side, scriptable command-line terminal for Segger-RTT; Also the AI-ready command line for embedded validation.

A self-hosted workspace where people and AI agents work together

248 small Jev decisions for JavaScript and TypeScript: route, rerank, gate, grade, compare, and label text for agents, RAG, support, code re

A fast, native terminal built on Ghostty, for everyday work

Native compact Nix build output with full local logs, bounded diagnostics, and a flake interface

jevals, agent trace'lerini eval ve guardrail kontrollerinden tek istekte geçirerek production veya agent loop içinde değerlendirmeyi amaçlayan bir Python kütüphanesidir.

Neden yükseliyor · 7 günlük bir repoda 97 yıldız, 5 commit ve 4 farklı yazar erken ilgi ve katkı işareti. Tek istekte birden fazla kontrol, düşük maliyet ve agent loop'a uygun hız iddiaları dikkat çekiyor; performans ve maliyet rakamları README'de verilmiş, bağımsız doğrulama olarak kabul edilmemeli.

Jev-Mem: System-One Controlled Agentic Memory

Declarative interactive HTML renderer and Page Spec runtime for AgentKit

Run Claude Code and Codex agents as a team: long-lived sessions, a durable mailbox, human approval gates and per-task token accounting, sche

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

Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents