novuhq/novu
Novu is an open-source communication infrastructure that unifies notification delivery across various channels for both products and AI agents, offering a single API for multi-channel messaging, wo...
Awesome AI Agents › Communication Protocols
agent-network-protocol/AgentNetworkProtocol
AgentNetworkProtocol (ANP) is an open-source communication protocol designed specifically for intelligent agents, aiming to become the foundational HTTP equivalent for the emerging Agentic Web era. The project envisions creating an open, secure, and efficient collaboration network that enables billions of intelligent agents to connect and interact seamlessly. ANP addresses critical challenges in agent networks, including interconnectivity to break down data silos, native interfaces that allow AI to interact with digital environments through APIs and communication protocols rather than mimicking human internet use, and efficient collaboration through automatic organization and negotiation among agents. The protocol architecture of ANP is layered and robust. It includes an Identity and Encrypted Communication Layer based on the W3C Decentralized Identifiers (DID) specification, which provides decentralized authentication and end-to-end encrypted communication without relying on centralized systems. The Meta-Protocol Layer facilitates negotiation of communication protocols between agents, enabling self-organizing and self-negotiating networks. The Application Protocol Layer uses semantic web standards to allow agents to describe their capabilities and manage supported protocols effectively. ANP is actively developed with an open-source implementation available at the AgentConnect repository. The project provides extensive documentation, including a beginner-friendly getting started guide, a technical white paper, and detailed specifications for decentralized authentication (did:wba), encrypted communication, meta-protocol negotiation, agent description, discovery, and messaging protocols. These specifications emphasize security, decentralization, and interoperability. The project also offers insightful blogs discussing the agent network concept, comparisons with other protocols, security principles, and technical approaches to AI-internet interaction. ANP aims to build a new digital infrastructure tailored for AI agents, promoting a future where intelligent agents collaborate efficiently and securely across diverse platforms and environments.
https://github.com/agent-network-protocol/AgentNetworkProtocol
Also in Communication Protocols
google-agentic-commerce/AP2
Agent Payments Protocol (AP2) is a project focused on creating a secure and interoperable protocol for AI-driven payments, providing code samples and demos to facilitate development and integration.
a2aproject/a2a-samples
Official sample repository for the Agent2Agent (A2A) protocol, with runnable agent and host implementations in Python, Go, Java, JavaScript and C# built on the official A2A SDKs.
universal-tool-calling-protocol/python-utcp
Python reference implementation of the Universal Tool Calling Protocol, a plugin-based standard that lets agents discover and call existing tools directly over HTTP, SSE, CLI and other transports.
google-agentic-commerce/a2a-x402
Specification and reference libraries for the x402 extension to the Agent-to-Agent protocol, letting one agent charge another for a service through on-chain stablecoin payments in a three-message flow.
a2aproject/a2a-java
a2a-java is the official Java SDK for the Agent2Agent protocol, providing client and server modules that expose agentic Java applications over JSON-RPC, gRPC and REST transports.
a2aproject/a2a-go
Go SDK for the Agent2Agent (A2A) protocol that provides client and server libraries for exposing and consuming agents over gRPC, REST and JSON-RPC, together with a CLI for discovery and messaging.
thu-nics/C2C
Research implementation of Cache-to-Cache, a method that lets two large language models exchange semantics directly by projecting and fusing their KV-caches instead of passing generated text between them.