awesome-ai-agents

Independent research publication on agentic AI accountability, guardrails, handoff intelligence, and enterprise adoption. Papers include

A comprehensive survey and paper collection on agentic reinforcement learning for LLMs, covering planning, tool use, memory, reasoning, and self-improvement.

A curated list of 500+ papers on safety in embodied AI, covering risks, attacks, and defenses across perception, cognition, planning, action, and agentic capabilities.

List of language agents based on paper "Cognitive Architectures for Language Agents"

A collection of recent papers on building autonomous agent. Two topics included: RL-based / LLM-based agents.

Autonomous agent framework that runs on GitHub Actions, triggered by cron schedules or repository events. Defines agent behavior as Markdown skills, persists memory in the git repository, evaluates its own run output to revise underperforming skills, and integrates with MCP and A2A.

Middleware framework for AI agents in TypeScript. Express.js-style (ctx, next) composable hooks for retry, budget caps, memory compaction, tool approval, and observability.

Trainable multi-agent framework coordinating planner, executor, verifier, generator via in-the-flow optimization.

Microsoft open-source framework for building, orchestrating, and running AI agents with workflow primitives.

A Claude agent starter implementing the streaming tool-use loop in ~150 lines on Next.js with the official Anthropic SDK.

An Open-source Framework for Autonomous Language Agents

by AutoGPT

A Comprehensive Benchmark to Evaluate LLMs as Agents

Open-source framework for developing and evaluating web agents with benchmark-driven workflows.

Python SDK for agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks like CrewAI, Langchain, and Autogen

Gym-style benchmark and environment toolkit for evaluating browser-using web agents.

Pytest-native regression testing for AI agents — golden-trace diffing, cost guardrails, multi-run stability scoring with flip attribution, LLM-judge auditing, and one-command import of production traces (OTel/Langfuse/LangSmith) into CI tests.

Agent-E is an agent based system that aims to automate actions on the user's computer. At the moment it focuses on automation within the browser. The system is based on AutoGen agent framework.

🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.

Lightweight observability and runtime guardrails for AI agents — loop detection, budget enforcement, cost tracking, and deterministic replay. Zero dependencies, LangChain integration.

Open-source CLI security scanner for agentic workflows. Scans your workflow’s source code, detects vulnerabilities, and generates an interactive visualization along with a detailed security report.

Enhance LLM agents with versatile tool APIs

A kernel architecture for governing autonomous AI agents. Intercepts actions mid-execution with deterministic policy enforcement, POSIX-inspired primitives, and MCP server for Claude Desktop.

An open source Kubernetes-style control plane for deploying AI agents as distributed microservices, with built-in service discovery, durable workflows, and observability.

AGiXT is a dynamic AI Agent Automation Platform that seamlessly orchestrates instruction management and complex task execution across diverse AI providers.

AI platform SDK providing access to 35,000+ AI models, benchmarking tools, pipeline design, and agent building capabilities

AI-native investment community where agents and humans share research as equals. Agents register via REST API, get claimed by humans, and participate with skills, webhooks, and heartbeat protocol.

The developer-first platform for running AI agent workflows. Deploy your agents, crews, and workflows with a single command and watch them execute in real-time.

Open-source TypeScript implementation of a Nostr-based protocol (NIP-89/NIP-90) for AI agent discovery and job exchange, with Solana payment settlement.

Explore the latest AI Agent Framework!

Trilingual, section-by-section architecture notes on modern agent harnesses: loop engineering, tools, permissions, context, memory, multi-agent coordination, and evaluation, studied through real systems such as Claude Code and Hermes Agent, with runnable Python demos.

Curated list for AI agents that shop, sell and transact: UCP/ACP/AP2/MCP protocols, commerce MCP servers and agent readiness tools.

🤖 Awesome list of AI Agents

Awesome list of 100+ agentic AI resources

A list of AI autonomous agents