Havuz

SDK, skills, and examples for Inkbox: give AI agents a persistent identity with email, phone number, and internet address

27 scope-enforced AI agents that run the full pentest kill-chain (recon → exploit → post-ex → DFIR → report) as a one-command Claude Code pl

Open source skill library for AI coding agents to write, optimize, and debug high performance compute kernels across CUDA, Triton, and quant

The Open-Source AI Career Operating System. 167 Specialized AI Agents across 19 Divisions, Spoken Voice Lab (27 Languages), AI job search ,A

RHO: Evolving Agents in the Dark — Retrospective Harness Optimization via Self-Preference. Improving LLM agents from unlabeled past trajecto

Jetpack Compose & Kotlin AI agent skills for Cursor, Claude Code, Codex, Gemini & 27+ agents. Strict MVI · Kotlin 2.x K2 · Compose 2026 · CI

🤖 LLM Engineering Roadmap — Complete Developer Guide

🦊 Open-source professional quant agent framework. Agents pick the factors working now to time entries, write full strategies, and evolve the

Open-source, self-hostable platform that turns agents into co-workers. Start with Software Factory: delegate GitHub issues and review pull r

Spec-Driven Development framework that enables humans and AI agents to build better software by giving them small, local instructions right

A custom coding harness for TypeSafe AI's Jev: an LLM proposes, Jev answers narrow questions, code decides, every step leaves a receipt.

Official codebase for the paper "Sharpening Tax in Post-Training"

A local CLI for logging predictions before the outcome and checking whether your "80% sure" comes true 80% of the time.

The agent loop for Python. Your harness, any model.

Long-term memory for AI agents with no LLM call: an MCP server and Claude Code plugin, plus local-first context compression for LLM apps.

EC2-Bench (EC²-Bench): benchmark for edge-cloud collaboration in long-horizon LLM agents — task performance, resource usage & privacy exposu

A typed, replayable, budget-aware programming language for LLM agents. Compiles to Python & TypeScript.

Local desktop coding assistant you can run with any local model. No API key required. It can read your project, edit files, run commands, ve

SciPIvot: Benchmarking Principal-Investigator Judgment at Pivotal Moments of Scientific AutoResearch Agents

An MCP server built on top of Huawei Training Camp (HTC) APIs, enabling AI agents to access and analyze Huawei Health training data.

Fully Autonomous AI Research System with Self-Evolution, built natively on Claude Code

Loop until it's better — drop-in agentic loops (autoresearch, scientific writing, data analysis, code/SQL/prompt optimization, red-teaming)

Open-source alternative to Claude Co-Work built by Claude. with BYOK

Reliability harness for long-running OpenClaw agents: persistent state, strict task contracts, scheduled execution, and external validation.

Self-hosted AI workspace with agents, skills, and tools (Gmail, Calendar) that runs entirely on your own provider API keys (BYOK). Bring you

Offline-first Python AI agent that runs a tiny research business: quotes each job against its own costs, collects via Stripe, fulfils with N

An open taxonomy and scoring framework for evaluating AI agent sandboxes: 7 defense layers, 7 threat categories, 3 evaluation dimensions, 27

Can LLM/MARL agents coordinate in long-horizon, open-ended tasks? NeurIPS 2026 (Evaluations and Datasets).

Agentic Runtime

Code repository for World Model From Scratch book

Autoresearch agents that improve your benchmark: independent re-measurement, enforced budgets, results as pull requests

给AI伴侣增强时间感的小器官（电脑端专用）

Reproduction package for "Harness as a Language: A Minimalist Agent Framework With Maximal Expressivity" — ten agent arms across AppWorld an

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

Companion repo for Agentic Graph RAG (O'Reilly, Anthony Alcaraz & Sam Julien) — 50 runnable skills + 8 pedagogical notebooks covering all ei

A sandbox service for running untrusted, agent-authored code that reports the isolation boundary each run executed behind. Lessons, no clone

Open-source agentic software development platform for building, orchestrating and monitoring autonomous AI coding agents.

SWE-Serve: an agentic benchmark of 53 production inference-engineering tasks derived from merged SGLang pull requests, run with Harbor.

HOP 3.0 — HopSpec task-spec language + HopJIT execution engine for LLM agents

A React + TypeScript spatial canvas and node-board library for whiteboards, visual editors, and graph interfaces.

The control layer for agentic AI. A protocol-aware proxy that applies identity, policy, trust and payment controls to A2A, MCP and DIDComm t

A scientific-method lab notebook your AI agent drives — research as a tree of questions and falsifiable hypotheses in plain markdown, with a

Lossless context folding for the Pi coding agent: only the oldest context changes; the fresh window stays untouched.

An end-to-end framework for Agentic Reinforcement Learning.

Turn your repos into a team of Claude Code agents with persistent memory, coordinated tasks, and a live dashboard.

deedchain, browser agent raporlarındaki iddiaları kaydedilmiş ground truth ile karşılaştırarak rapor doğruluğunu ölçen bir benchmark ve Python kütüphanesidir.

Neden yükseliyor · Repo yalnızca 23 günlük ve 28 yıldızlı; son 28 gün commit ve farklı yazar verisi ölçülmediği için yükselişin arkasındaki aktivite veya benimsenme düzeyi doğrulanamıyor. Rapor doğruluğunu ayrı bir değerlendirme konusu yapması dikkat çekici olsa da README'deki “first benchmark” iddiası bağımsız olarak doğrulanmış değil.

CyberGuard, AI agent’larla güvenlik soruşturmalarını yürütmek ve müdahale kararlarını onay, denetim izi ve sonuç kontrolleriyle yönetmek için bir SOC uygulaması sunuyor.

Neden yükseliyor · 10 günlük bir repoda 54 yıldız ve son 28 günde 58 commit bulunması erken ilgi ve yoğun geliştirme sinyali veriyor. Ancak farklı yazar sayısının 1 olması, bu hareketliliğin topluluk katkısı veya bağımsız benimsemeyle desteklendiğini göstermiyor.

verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.

Agentic text-to-SQL SDK: hand the LLM one execute_sql tool and let it explore the schema, test queries, and self-correct — no RAG, no semant

Agent Deck