Havuz

Free & Open Source AI Agent — Self-Hosted, Self-Learning, No Vendor API. Qwen2.5-7B + LoRA. 35 tools. Built on Islamic Epistemology (IHOS).

Pi agent for Android — sideload the APK and run a Pi coding agent on your phone. Single-APK, on-device Termux runtime, Claude Pro/Max OAuth,

A video-call avatar frontend for your local LLM agent (Hermes/OpenClaw) — replaces the messenger chat and serves the agent's exact slash-com

SOUL.md governance framework for Hermes Agent — structured memory, skill management, and operational rules

纯 Kotlin 原生 Android AI 智能体：多模型接入、子智能体协作、在线插件市场、技能自动引用、root/Ubuntu 终端、长期记忆

🪼 A personal AI assistant that lives in Telegram — every topic is its own isolated Claude Code agent with layered persistent memory and proa

OpenClaw 源码深度研究：27 章技术主线 + 2 月以来 PR 演进 + fork/变种生态 + 缺陷与 roadmap（2026-04 snapshot）

Cross-platform Rust computer-use MCP server for macOS and Windows: AX/UIA, screenshots, OCR, pointer and keyboard input, clipboard, stdio, a

A menu-bar macOS agent loop: every N minutes it asks Claude/Codex what's overloading your Mac and suggests one-click, risk-checked fixes.

A coding assistant built on persistent multi-agent teams: durable members that discuss, chat, and re-sync continuously — Discuss -> Code in

The runtime substrate for agentic systems — one Go binary, six LLM providers, MCP-native, configurable as a managed sandbox or full agentic

生产级 3GPP 5G 规范 RAG Agent：自然语言提问，回答带段落级原文引用 + 严格 grounding，覆盖 Rel-18/19 全部 TS。

Mythos-class autonomous vulnerability discovery from any strong coding model (Opus 4.7, GPT-5.5, DeepSeek V4, Gemini). Long-running, email-d

Experimental DSH Minimal trajectory-steering adapters for DeepSeek V4 Pro on Pi, Oh My Pi, and OpenCode V2. 推理轨迹引导实验。

Hybrid vector + knowledge graph RAG system for enterprise documents. Fuses Pinecone dense vectors, BM25 sparse reranking, and Neo4j multi-ho

Minimal surface, complete architecture: a small, local-first coding-agent harness that re-derives the invariants of DeepSeek Harness with fe

铁锈战争单机 LLM 指挥链：javaagent 实时取数 + Python 分层多 agent 决策（纯单机研究项目，严禁用于多人对战）（后期架构大改变，现阶段有些狗屎）

The open-source alternative to Suno and ElevenLabs Music. Natural language music composition, run locally, own everything.

Lawgent — an open-source legal AI agent. Grounded, multi-jurisdiction (CN · US · EU · ET · ICAO), citation-verified: every pinpoint round-tr

Cord — distributed agent fabric for LLMs, MCP servers and AI agents. Discover any LLM CLI / HTTP backend / MCP server across machines via na

An open source AI roleplay agent harness. One click builds an autonomous character agent that creates with you. Websites, code, music, video

MCP server that gives text-only LLM coding agents vision — analyze images via any multimodal model (mimo, Claude, Gemini, OpenAI-compatible)

Atria Dawn Preview, araştırma ve mühendislik görevlerinde araç kullanımı ve çok adımlı iş tamamlama için sunulan bir agentic modeldir.

Neden yükseliyor · 15 günde 555 yıldız, son 28 günde 43 commit ve 6 farklı yazar kayda değer erken ilgi ve aktivite gösteriyor. İlginin kesin nedeni bilinmiyor; yeni model yayını ve README'deki benchmark iddiaları dikkat çekmiş olabilir, ancak bu veriler tek başına model kalitesini kanıtlamaz.

An AI troubleshooting workbench for internal systems: it integrates ticketing and alerting systems via HTTP APIs and webhooks, connects code

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?

Local-first, model-agnostic LLM agent-harness research. Restart planned after a hiatus; existing code and historical claims await revalidati

Self-hosted AI agent living inside Zalo (personal account). Multi-account, 15 tools + external MCP servers, web dashboard, scheduler, durabl

dots, OpenRouter üzerinden seçilen bir modelin Firefox tabanlı bir tarayıcıyı kullanarak web görevlerini yerine getirmesini sağlayan bir AI agent arayüzüdür.

Neden yükseliyor · Bir günde 1.957 yıldız alması dikkat çekici; anti-detect browser ile AI agent kullanımını tek bir kurulabilir arayüzde birleştirme vaadi ilgi çekmiş olabilir. Ancak yıldız sayısı çok yeni bir repo için benimsenme veya teknik başarının kanıtı değil; son 28 günde 4 commit ve 1 farklı yazar verisi de bağımsız geliştirme etkinliğinin henüz sınırlı olduğunu gösteriyor.