Python agents

Pebblo enables developers to safely load data and promote their Gen AI app to deployment

Code for Paper: Autonomous Evaluation and Refinement of Digital Agents [COLM 2024]

Model Context Protocol server to allow for reading and writing from Pinecone. Rudimentary RAG

[EMNLP 2024] OneGen: Efficient One-Pass Unified Generation and Retrieval for LLMs.

We present the first systematic study on the scaling property of raw agents instantiated by LLMs. We find that performance scales with the increase in the number of agents, using the simple(st) way of sampling and voting. Our method is called Agent Forest, as a tribute to the classic Random Forest.

[EMNLP2025]Official implementation: Agent-style vision question answer in Autonomous Driving!

A knowledge source about TTPs used to target GenAI-based systems, copilots and agents

Ultra-Lightweight, Pure Python Multimodal Agent.

Self-hosted AI assistant with tool use, multi-agent orchestration, coding copilot and a lightweight Flask + vanilla JS stack.

桌面 AI 副屏 Skill：M5Paper 墨水屏 + AI Agent 推送 widget。装 Skill 后由 Agent 自动引导烧固件、配 Wi-Fi、推送内容。本地运行、无云依赖。

Practices, protocols, and skills for AI-driven software development. Skills and safety hooks for Claude Code, Codex, OpenCode, Cursor, Antigravity, and any agent supporting the Agent Skills standard.

OpenSRE — the memory-first AI SRE. Self-hosted incident investigation with episodic memory, knowledge graph, web console, Slack & Teams. opensre.in

Local-first scientific literature RAG with document-count thresholds, model-free worksheets, frozen evidence, and persistent human review.

Auditing AI agents: a curated list of papers, tools, datasets, benchmarks, and standards covering reliability, monitoring, failure attribution, and decision records.

gradio region:us

transformers safetensors llama text-generation conversational arxiv:2403.04652

transformers pytorch safetensors llama text-generation llama-2

Local-first agent workspace for knowledge, tasks, documents and verifiable workflows · Локальная агентная система для знаний, задач и проверяемых процессов · t.me/romarayt

Agent skills that fact-check the internet: claim-by-claim verification with sources and a 0-10 BS score for any YouTube video, article, tweet, or PDF

[EMNLP 2026] AgentCompass is an extensible open-source evaluation infrastructure for systematically assessing LLM/VLM agent capabilities.

Turn scattered notes, docs and transcripts into a queryable Markdown wiki — an LLM knowledge-base compiler with MCP access, no embeddings, self-hosted.

A multi-agent research laboratory for data distillation and long evolutionary collaborative reasoning.

Convert the given plain text to MySQL query by ChatGPT

A mcp server that bridges the Model Context Protocol (MCP) with the Agent-to-Agent (A2A) protocol, enabling MCP-compatible AI assistants (like Claude) to seamlessly interact with A2A agents.

An OpenaAIGym-based framework allowing to test hybrid approaches (RL + path planning) for multi-UAV systems that are supposed to provide smart services.

This is the code for Siraj Raval's Video on Building an Autonomous Polymarket Agent

Vibe coding on your mobile phone. A modular, self-evolving AI-powered system for Termux in your phone, designed to control the device and automate complex tasks, create apps in seconds. If you can imagine it, the AI Agent can build it, execute it, and expand the system's own capabilities.

transformers safetensors llama text-generation license:llama2 text-generation-inference

transformers safetensors deepseek_v2 text-generation conversational custom_code

transformers safetensors qwen2 text-generation code codeqwen

gguf quantized moe vlm vision agentic

协调探索、执行和独立评审，让复杂任务按明确职责并行推进，由主 Agent 统一验收。

轻量级多智能体协作系统 —— AI Agent 开发入门实战项目

Implementing Skills functionality for your agents

🪶 Lightweight OpenAI drop-in replacement for Kubernetes

Let's be honest, who really knows how to use ffmpeg. Its that tool that is so helpful but not needed enough to justify learning all of its inner workings. The days of scrolling through stackoverflow are over, with llmpeg. This CLI tool interfaces with an OpenAI LLM to convert your plain text commands into OS/platform specific ffmpeg commands.

see live demo of chatbot. follow the link

AI-powered analysis engine for Argentine government open data. Multi-agent pipeline (LangGraph) with 10 data connectors, NL2SQL, semantic caching, and real-time streaming. Built with FastAPI, PostgreSQL + pgvector, Celery, and Gemini/Claude LLMs.

[NeurIPS 2025] ClinicalLab: Aligning Agents for Multi-Departmental Clinical Diagnostics in the Real World

Run a <400ms latency Voice Agent on just 4GB VRAM. Fully offline, no API keys required. Optimized for GTX 1650 and edge robotics with zero-copy inference. (Apache 2.0)

A collection of specialized agent skills for AI infrastructure development, enabling Claude Code to write, optimize, and debug high-performance systems.

A private, modular node based AI workspace for thinking and creation. Seamlessly connect to Ollama, OpenAI, Gemini, Anthropic, and Groq models in one unified, node-based interface for reasoning, visualization, and exploration — all running locally or via API.

Open, MIT context format for AI coding agents: centralized .aigx/ rules + a per-file boundary index that targets the right rules, forbidden imports, and gotchas to each file. The only context format validated in a controlled benchmark. Tool-agnostic; nothing injected into your source.

transformers safetensors llama text-generation conversational license:other

Terminal-first, knowledge-grounded multi-agent software delivery pipeline: scope requirements, implement changes, run tests, and gate pull requests with deterministic QA and ensemble code review.

Harnessing the Memory Power of the Camelids

The multi-agent version of TORCS for developing control algorithms for fully autonomous driving in the cluttered, multi-agent settings of everyday life.

Create an AI that chats like you

Implementation of PALI3 from the paper PALI-3 VISION LANGUAGE MODELS: SMALLER, FASTER, STRONGER"

Resumable, traceable AI coding — decisions and history stay with the project

Machine Learning and having it Deep and Structured (MLDS) in 2018 spring

tensorflow reinforcement learning agents for OpenAI gym environments

Openai-style, fast & lightweight local language model inference w/ documents

An open solution for AI-powered photorealistic digital humans.

🤖Free Agent Line Bot with Web Search, Google Image Search, Image Generator, Video Generator...

CRM Data Q&A Agent - Advanced RAG with NL2SQL over Salesforce Data

Custom tools to enhance your Open-Webui Experience! 🚀

Multi-Agent Retrieval-Augmented Generation (RAG) Customer Support System using Python, LangChain, and LangGraph.

A shared repository for skills. Intended to be used with Letta Code, Claude Code, Codex CLI, and other agents that support skills.

[ICLR/AAAI/KDD/EMNLP2026] Open-Source LLM-Based Data Analysis Agents

AI Agency Command Center for Claude Code — orchestrates 5 AI teams (Marketing, Sales, Legal, Reputation, GEO/SEO) into a unified zero-employee agency. 9 skills, 5 parallel agents, composite scoring, and professional PDF reports.

A collection of AI Skills specialized in Django Backend development

50 QA and test-automation skills for Claude Code, Codex, Cursor, and any Agent Skills Standard runtime.

transformers pytorch chatts feature-extraction text-generation conversational

AI coding agent with one Python core and three front-ends — headless CLI, Textual TUI, and an Electron desktop. Works with any OpenAI-compatible API, with risk-tiered permissions, event-sourced replayable sessions, and a fail-closed OS-level sandbox.

基于seq2seq模型的简单对话系统的tf实现，具有embedding、attention、beam_search等功能，数据集是Cornell Movie Dialogs

SmartPlay is a benchmark for Large Language Models (LLMs). Uses a variety of games to test various important LLM capabilities as agents. SmartPlay is designed to be easy to use, and to support future development of LLMs.

CrawlAI RAG is an AI-powered website intelligence platform that allows users to crawl entire websites, index their content, and ask natural-language questions using Retrieval-Augmented Generation (RAG). It transforms static websites into queryable knowledge bases.

Chat with GPT LLMs over voice, UI & terminal, all with access to the internet. Powered by OpenAI.

multi agent RL for traffic light control in Sumo using distributed PPO

Tool that just makes your open source project better using LLM agents

Awesome ModelContextProtocol resources - A curated list of MCP resources

"Generative AI in Action" book's code repository

Conversational Health Agents: A Personalized LLM-powered Agent Framework

A multi-agent quadruped environment, supporting learning of locomotion control or merely high-level planning.

AI agent to automatically generate and post short videos

Benchmark for LLM-based Agents in Computational Biology

Agent Skills for the Venice.ai API. One folder per surface area, each with a SKILL.md for agent runtimes (Cursor, Claude, Codex, etc.).

[ICLR 2026] Beyond a Million Tokens: Benchmarking and Enhancing Long-Term Memory in LLMs

多Agent智能教育与个性化学习系统 | 5-Agent Mesh+事件驱动 | Java/Go/Python三语言实现 | 面试项目 | 八股文+STAR+简历

MCP server that lets Claude Code and other AI agents read and search large PDFs, one file or a whole folder: agentic RAG with hybrid semantic + keyword search, selective page reads, tables, images, OCR, chart data, and multi-column/CJK layouts.

Eva01 is NOT an assistant. She is an AI being with her own mind, feelings, and intrinsic drives. Multimodal, Modular design. Built-in voice & face recognition. Plug'n play tools. Compatible with ChatGPT, Claude, Deepseek, Gemini, Grok, and Ollama. Explore the possibilities of Human-AI Interaction.

Clean-room, model-agnostic harness for Claude Code-class coding agents

GateMem: a benchmark and evaluation toolkit for memory governance in multi-principal shared-memory LLM agents.

An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch-1 LLM decode, paper: https://arxiv.org/abs/2606.09682

A context harness for AI agents: all your scattered context — code, memory, docs, databases, SaaS — in one searchable, browsable, file-like interface.

transformers safetensors lfm2 text-generation liquid edge

Agent Execution Partnership AEE is an open-source control plane that ensures every AI agent action is authorized before it runs, observable while it runs, and verifiable after it completes.

When you can't trust yourself with your code base, trust Arbiter.

This project enhances the construction of RAG applications by addressing challenges, improving accessibility, scalability, and managing data and user access. It uses Django, Llamaindex, and Google Drive for effective database management.

Seed, Code, Harvest: Grow Your Own App with Tree of Thoughts!

Proximal Policy Optimization (PPO) algorithm for Contra

[EMNLP'24] EHRAgent: Code Empowers Large Language Models for Complex Tabular Reasoning on Electronic Health Records

Official Implementation of "Psychologically Enhanced AI Agents"

LLM-powered assistant for creating, editing, and interpreting business process diagrams

KiroaaS (Kiro as a Service) Your Gateway to Kiro. Expose Kiro's models to any application via standard APIs. Power your tools with Kiro's intelligence.

Unreal Engine MCP Server: Control UE5 with Claude & AI Agents. Supports Python and C++ for custom tool development.

Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses | 500+ Papers | Perception, Cognition, Planning, Interaction, Agentic System