Havuz

基于 Go 与 Eino 的多模态视频解读 Agent：阅读转写文本/字幕、按需查看真实画面，将 B站和 YouTube 视频整理为图文文章，支持围绕原视频持续问答。

The coding agent you don’t have to watch. Start a task and walk away. AFK builds the feature, verifies its own work, and texts you when it’s

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

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

From Agent to Agency — 一个 AI 做不了的事，一群 AI 可以。Multi-Agent Collaboration Platform.

A JAX transport-economy learning environment for route planning, cargo flow, financing, and replayable agent benchmarks.

Pinax is a small Go CLI that crawls a public documentation site, indexes its structure, and exposes it to any Model Context Protocol client

Run coding agents unattended. Agents write the code. Axocoatl writes the receipt. Claude Code, Codex or a local model in a sandbox; the host

Karpathy's autoresearch loop, applied to outbound. Agent-operated GTM that scores replies AND silent self-serve across email, LinkedIn, and

Stack-agnostic agentic-programming workflow skills + documentation scaffold

CounterStrikeSharp plugin for CS2 weapon skins, knives, gloves, and agents.

Coding agent harness, achieved SWE Bench Lite resolve rate of 52.4%

Hugging Face recipes: run OCR, transcription, classification, embeddings and more over a Hub dataset in one command, locally or on HF Jobs.

Counterfactual replay and causal attribution over LLM-agent trajectories: find which step actually caused a failure by intervening on it.

Evolving agent harnesses: a research program on how far N orchestrated calls of a small model can rival a frontier model. We evolve the harn

Free guerrilla growth tactics for startups, the kind your competitors won't come up with on their own. Runs in Claude Code or Codex: competi

Frappe Flow — native AI agents, tools, and triggers for Frappe

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

Your agent team, on your machine. Open-source harness for multi-user Claude agents.

Open-source visual ai builder that lets you design AI visual ai agents workflows by dragging and dropping nodes on a canvas, and then compil

Agentic pipeline: product photo → articulated 3D asset (USD) for robotic simulation

Hardware-in-the-loop benchmark for agentic embedded AI deployment. Agents iteratively edit model or firmware artifacts, the framework compil

Run Claude Code dynamic workflows on a local Codex (GPT) backend, plus an interactive run viewer

A library for modelling 3D objects with mathematical primitives

AI agent skill for creating cinematic product launch videos with Remotion

Emotional long-term memory framework for AI agents and companions

Munder Difflin, mevcut terminal tabanlı coding agent’larını yerel bir masaüstü uygulamasında koordine eden multi-agent harness’tir.

Neden yükseliyor · Çoklu agent iş akışlarına yönelik ilgiyle, zaten kullanılan terminal CLI’larını tek bir arayüzde birleştirme vaadi örtüşüyor. 119 günde 8.016 yıldız ve son 28 günde 177 commit ile 7 farklı yazar kayda değer ilgi ve geliştirme aktivitesi gösteriyor; README’deki trend rozetleri ise proje beyanı olduğundan tek başına benimseme kanıtı sayılmaz.

A generalist autonomous research agent — runs experiments, researches, and iteratively optimizes, autonomously.

Reusable AI tools for the Laravle AI SDK

Open source coding agent workspace, an agent-first desktop app that runs locally. alternative to Cursor and Codex.

AI agent with zero dependencies, bring your own bun

211 free visual cards about large language models, from tokenization to deployment. Full collection: 376 cards.

Tandem lets you connect and control any coding agent—including Claude, Codex, or Hermes—across different devices. It allows all your agents

🧠 A council of AI models for the decisions that aren't no-brainers — they answer in parallel, debate to consensus, or get judged to a ver

Turn any supported coding-agent harness into a full game-dev studio - 55 specialized AI agents and 182 skills for Godot, Unity, Unreal, Thre

AgentGuard: Zero-Trust Security Foundation for AI Agents

Knack CLI: agent-friendly skill operations. Self-host on GitHub or use Knack Cloud.

MergeOS — bounty marketplace, ledger, scan, and agent protocol platform

Open-source tools for AWS DevOps Agent - extend DevOps Agent with ready-to-use skills, custom agents, and other tools, for incident response

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

Let AI Agents control your real firefox browser: fill forms, authenticate, automate actions, web development; all via token-efficient, easy

The first benchmark that asks code agents to build agents.

中文公司调研 Agent

Skills for agents

A small, typed, hookable agent loop. Provider-agnostic, sandbox-agnostic, tooling-agnostic. Battle tested on clarkchat.com

ADHD, coding agents için bağımsız fikir dalları üretip bunları puanlama ve eleme aşamalarından geçiren bir ideation skill’idir.

Neden yükseliyor · 125 günde 4290 yıldız ve The New Stack haberi belirgin ilgiye işaret ediyor; ancak README’deki eval örneği tek bir problem ve LLM judge karşılaştırmasına dayanıyor, bu yüzden etkinlik iddialarını sınırlı kanıt olarak değerlendirmek gerekir.

The missing bridge between your ML models and your AI agents.

Runtime safety net for LLM agents. Detects token spirals, kills doomed tasks early, tells you exactly why. Rust core, Python SDK. pip instal

Automated auditing pipeline for LLM and agent benchmarks — surfaces task ambiguity, environment conflicts, and evaluation bugs.