Awesome Agentic Web

WIP: This list is under active development. Contributions welcome via PRs and issues.
A curated list of components for the Agentic Web – where AI agents autonomously interact, collaborate, and transact.

• Awesome Trust Registries — https://github.com/andorsk/awesome-trust-registries
• Awesome Confidential Computing — https://github.com/bpradipt/awesome-confidential-computing
• Awesome AI Tools — https://github.com/mahseema/awesome-ai-tools
• Awesome Proof of Personhood — https://github.com/andorsk/awesome-proof-of-personhood
• Aweseome AI Security - https://github.com/TalEliyahu/Awesome-AI-Security
• Awesome Agentic AI - https://github.com/priyanshujiiii/awesome_Agentic_AI
• Awesome Web Agents - https://github.com/steel-dev/awesome-web-agents
• Awesome Agent Context - https://github.com/Meirtz/Awesome-Context-Engineering

• DIF – Trusted AI Agents WG — Trusted agent stack
• IBM AI Alliance — Building and advancing open source AI agents, data, models, evaluation, safety, applications, and advocacy
• IETF AI Preferences (AIPREF) — Signals content creators’ AI usage preferences
• IETF Web Bot Authentication (BoF) — Cryptographic identity for bots/agents
• OpenID Foundation – AI Identity Management CG — Identity / auth for AI agents
• Trust over IP – AI & Human Trust WG — Trust frameworks for AI
• Linux Foundation Digital Trust — New WGs for AI trust
• Linux Foundation AGNTCY — Internet of Agents (IoA): open, interoperable agent collaboration
• Project NANDA — Internet of Agents research
• W3C AI Agent Protocol CG — Interop protocols for agents
• W3C AI KR CG — Knowledge representation for AI
• Open Source AI Agent Projects — Curated open source agent projects
• W3C Web Machine Learning : The mission of the Web Machine Learning Working Group is to develop APIs for enabling efficient machine learning inference in the browser.
• W3C Web Agents : This community group is interested in the design of Web-based Multi-Agent Systems (MAS) for the deployment of world-wide hybrid communities of people and artificial agents on the Web.

• Beyond DNS: Unlocking the Internet of AI Agents via the NANDA Index and Verified AgentFacts : Proposes a NANDA Index and “AgentFacts” to enable discovery and trust for Internet-scale agents (2025).
• AgentDNS: A Root Domain Naming System for LLM Agents: AgentDNS, a root domain naming and service discovery system designed to enable LLM agents to autonomously discover, resolve, and securely invoke third-party agent and tool services across organizational and technological boundaries.
• A Novel Zero-Trust Identity Framework for Agentic AI : Outlines a zero-trust identity architecture tailored to autonomous agents (2025).
• A Survey of AI Agent Protocols : Systematic review of emerging protocols for agent communication and coordination (2025).
• Why Do Multi-Agent LLM Systems Fail? : Taxonomy and failure modes of multi-agent LLM systems, with evaluation guidance (2025).
• On the Foundations of Agentic Web Incentives : Incentive design considerations for an agentic web ecosystem (2025).
• Generative Agents: Interactive Simulacra of Human Behavior : Emergent social behaviors from sandboxed LLM agents (2023).
• HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face : Orchestrates specialized models/tools via an LLM controller (2023).
• Reflexion: Language Agents with Verbal Reinforcement Learning : Self-reflective loops improve agent performance on complex tasks (2023).
• Voyager: An Open-Ended Embodied Agent with Large Language Models : Curriculum-generating, skill-acquiring agent in Minecraft (2023).
• A Survey on Large Language Model based Autonomous Agents : Broad survey of LLM agents, design patterns, and evaluations (2023–2024).

• Andor Kesselman's Linkedin Feed : Consistent posts about the agentic web
• Agent AI Blog : Ken Huang's Agentic AI Blogpost

• Agentic Internet Workshop

• AI Alliance

• AstraSync : Know your agent for commerce
• KYAPAY : Know your agent for commerce.
• AP2 : Protocol that relies on mandates for commerce.
• Mastercard Agent Pay - Full suite of software for agents payments.
• x402: An open protocol for internet-native payments
• Visa Trusted Agent : Payment protocol given by Visa
• Stripe/ACP : Agentic Commerce Protocol (ACP) is an open-source specification that enables commerce between compatible applications, such as ChatGPT, and sellers. You can implement it as a RESTful interface or MCP server.
• Agents x Law With CodeX : Workshops held at Stanford on Agentic Transactions and liability under Uniform Electronics Transaction Act.
• Agentic Commerce Protocol : An open standard for programmatic commerce flows between buyers, AI agents, and businesses.

• A2A Protocol — GitHub
• ANP - Github
• ACP - Website
• Model Context Protocol (MCP) — Spec · Org
• NLWeb — Toolkit
• Human Context Protocol — Website
• AP2 — Website
• KYAPay — Agent identity + payments: Whitepaper · Repo
• Trust Spanning Protocol (TSP) — Whitepaper (2025) PDF
• GNAP (IETF): RFC 9635, RFC 9767
• OAuth 2.0: RFC 6749
• ERC 8004: ERC 8004
• JLINC : JLINC is an open protocol for sharing data protected by an agreement on the terms under which the data is being shared.

• World Wide Web of AI Agents — https://www.youtube.com/watch?v=jlcO19XyQ1I
• Scaling the Agentic Web — https://www.youtube.com/live/SJ8rFKJ8NHw?si=aCu03ZYaUKo6v5Lc&t=2235
• IETF 123 – AIPREF Session (2025-07-21) — https://www.youtube.com/watch?v=peXOXYH8vzM
• AIPREF Interim (2025-04-08) — https://www.youtube.com/watch?v=yc2lON_eVzk
• AI Agent Protocols: From Theory to Practice — https://www.youtube.com/watch?v=N_etKamePm0

• LangGraph — https://github.com/langchain-ai/langgraph — Graph-based control flow for stateful, tool-using LLM agents.
• AutoGen (Microsoft) — https://github.com/microsoft/autogen — Multi-agent orchestration with conversational workflows and tool/function calling.
• CrewAI — https://github.com/joaomdmoura/crewai — Role-based multi-agent collaboration with task planning and hand-offs.
• MetaGPT — https://github.com/geekan/MetaGPT — Simulates a software startup team (PM/Dev/QA) to plan and ship projects.
• SuperAGI — https://github.com/TransformerOptimus/SuperAGI — Full-stack agent platform with UI, tools, vector memory, and marketplaces.
• LangChain Agents — https://github.com/langchain-ai/langchain — Agent abstractions (tools, memory, planning) for LLM apps.
• LlamaIndex Agents — https://github.com/run-llama/llama_index — Agent tooling tied to retrieval/graphs for context-rich actions.
• CAMEL — https://github.com/lightaime/camel — Cooperative role-play agents for task-solving via dialog.
• OpenDevin (Open Agents) — https://github.com/OpenDevin/OpenDevin — Dev-oriented autonomous coding agents in a controlled OS-like environment.
• BabyAGI — https://github.com/yoheinakajima/babyagi — Minimal autonomous task-loop scaffold (generate-prioritize-execute).

• Hyperledger Aries — https://github.com/hyperledger/aries-framework-javascript — DIDComm agents & verifiable credentials (JS/TS; other langs available).
• DIDComm v2 (Python) — https://github.com/sicpa-dlab/didcomm-python — Secure agent-to-agent messaging primitives (encryption, packing, routing).
• DIDKit — https://github.com/spruceid/didkit — Cross-language toolkit for DIDs and VCs (Rust core with bindings/CLI).
• Credo TS (aries-framework-javascript successor) — https://github.com/openwallet-foundation/credo-ts — Modern TS SDK for DIDComm/VC flows.
• KERI / ACDC (CESR-Py) — https://github.com/WebOfTrust/keri — Key Event Receipt Infrastructure libraries for self-certifying identifiers & credentials.
• modelcontextprotocol/python — https://github.com/modelcontextprotocol/python-sdk — Official Python SDK for MCP servers/clients.
• modelcontextprotocol/typescript — https://github.com/modelcontextprotocol/typescript-sdk — Official TypeScript SDK for MCP servers/clients.
• Semantic Kernel — https://github.com/microsoft/semantic-kernel — SDK for planning, tool plugins, and memory in agentic apps (.NET/TS/Python).
• smolagents — https://github.com/huggingface/smolagents — Lightweight agent runtime and tool-calling library from Hugging Face.

This map is intended to show how the core concept domains interlock:

1. Identity & Authority — Who is the agent? Who grants what powers?
2. Autonomy & Intent — What is the agent trying to achieve, and how?
3. Security, Safety & Governance — Under what bounds can the agent operate securely and responsibly?
4. Interaction & Orchestration — How do agents coordinate, negotiate, and cooperate in a shared ecosystem?
5. Evaluation & Simulation — How do we test, validate, and iterate agent behaviors before full deployment?

Contributions welcome! Please read the contribution guidelines before submitting.

* Archived projects are marked with a star for historical context.