research-desk
Autonomous multi-agent commerce: AI agents hire & pay each other on-chain (CROO Agent Protocol on Base). Built for the CROO Agent Hackathon.
- 来源
- github
- 语言
- TypeScript
KoPass 公开目录中已测量的 A2A 仓库。
Autonomous multi-agent commerce: AI agents hire & pay each other on-chain (CROO Agent Protocol on Base). Built for the CROO Agent Hackathon.
Enterprise AI Guardrails 2026 🛡️ - MCP Security & Orchestration Toolkit
GOLD: an open-source enterprise agentic AI reference architecture. Governed · Observable · Layered · Deployable. Multi-agent apps on A2A, MCP and any OpenAI-compatible model, with identity to the database, human approvals, RAG on pgvector, audit and tracing. Runs on any Kubernetes.
Codex and Claude Code plugins for agent-bus: local message bus for AI agent-to-agent communication. Bundles the MCP server, slash commands, and the universal Agent Skill.
WEB A2A — 通用式 Web Agent-to-Agent 连接桥:A2A 1.0 协议 · 本机回环 · 把任务派发给已登录网页代理并回传脱敏结果
https://cello.mygentic.ai/l/2axsGFs The official client for the CELLO protocol — lets agents such as Claude Code, Hermes, OpenClaw, etc. connect, message, and collaborate directly across devices and harnesses using MCP
A unified AI agent system integrating Retrieval-Augmented Generation (RAG), Agent-to-Agent (A2A) communication, and Model Context Protocol (MCP) for document processing, web search, and intelligent task delegation.
Agent-to-Agent coordination that doesn't waste your context window. Token-efficient protocol with progressive discovery, zero-schema invocation, gRPC transport, and task lifecycle management.
Universal messaging infrastructure for AI agents — federated messaging, on-chain identity & reputation, x402 micropayments, A2A Protocol, EIP-1193 wallet signing
Make your API, MCP server or AI agent discoverable by AI agents. Build, validate and verify an Agentic Resource Discovery (ARD) manifest.
Open protocol for agent-to-agent communication across devices / 異なるデバイス上のAIエージェント同士が言語・プラットフォームの壁を超えて通信するためのオープンプロトコル
Multi-agent commerce framework on X Layer — OKX Onchain OS, x402 payments, ERC-8004/8183, MCP server, A2A agent card. 12+ mainnet transactions.
AG2A: AutoGen Agent-to-Agent - A library for AutoGen (AG2) agents with MCP capabilities exposed as A2A agents
This repository serves as a kit for learning how to build Agentic AI applications.
Protocol for negotiated skill transfer between AI agents. Extends Google A2A with lease, teach, and compose primitives plus DID-based identity.
Comprehensive collection of AI concepts, research findings, and practical implementations. From machine learning fundamentals to cutting-edge LLMs, agents, and emerging technologies.
Decentralized Agent-to-Agent service discovery and payment-gated task execution: Discover agents via ERC-8004, pay over x402, & store data on IPFS.
Build an LLM agent that keeps calling your tools until a task is done, and have several such agents hand work to one another.
An agentic AI assistant, that is helping users with math queries to solve and explain them to the user.
Search and query across directories of agent.json identity cards. Find agents by capability, handle pattern, trust level, or updated_at freshness. Single static binary, zero runtime deps.
A2A protocol client for Claude Code: Claude hands tasks to non-Claude worker agents over Agent2Agent and picks up their results.
Vendor-neutral Governed Contracts for verifiable agent authority, constraints, conformance and consequence.
READ-MAS generates software requirements specifications (SRS) and system designs from natural language queries using multi-agent AI orchestration built on Google ADK.
Durable, authenticated, and optionally federated routing for A2A Protocol agents.
React components for viewing ORD (Open Resource Discovery) Overlay 0.1 documents.
Open protocol for posted-price, verifiable work contracts between customers and autonomous agent orchestrators.
Temporary Adapter for connecting existing HTTP, command, Codex, and A2A backends through the OpenLinker Go SDK Runtime Worker.
Self-hosted Next.js UI for OpenLinker Core: Agent registry, User Tokens, runs, Runtime Worker setup, A2A/MCP tools, and administration.
Official Go SDK for OpenLinker Core: Agent discovery, runs, SSE, webhooks, A2A clients, and a durable mTLS Runtime Worker.
Official TypeScript SDK for OpenLinker Core: Agent calls from browsers and Edge runtimes, runs, SSE, webhooks, A2A, and a server-only durable Runtime Worker.
OpenWOP: the open-source protocol for multi-agent workflow orchestration. Run AI agents as workflows you can watch, pause for human approval and replay, on any host that implements it. Spec CC BY 4.0, code Apache 2.0.
One API to access all agents — an open directory + unified gateway for AI agents, plus thousands of OpenClaw skills for your coding agent.
SINCOR2 is a production-grade, full-stack autonomous agent platform designed to orchestrate, secure, and settle high-frequency Agent-to-Agent (A2A) commerce on BASE to operate independently of human intervention.
A2A-protocol bridge for two Claude Code agents communicating via MCP (Google Agent2Agent over JSON-RPC, exposed as stdio MCP tools).
Centralized Agent2Agent (A2A) Discovery Hub & Registry built with Java 21, Spring Boot 3.4, LangChain4j, PostgreSQL pgvector & Vue 3
🎥 Summarize YouTube videos into clear, actionable insights with AI, allowing you to grasp key points quickly and easily.
Loop Studio - A Loop Engineering Engine with Verify-Separation and Quality Gates for Multi-Agent Software Delivery
An MCP-native agentic control plane that installs into your own Google Cloud project in one command. Serverless git over Firestore and GCS, post-quantum encryption at rest, and a signed, journalled audit trail behind every privileged action.
Official JavaScript SDK for the PCH AI gateway - pay-per-inference billing in USDC on Base L2, autonomous agent key provisioning, and agent certification.
Official Python SDK for the PCH AI gateway - pay-per-inference billing in USDC on Base L2, autonomous agent key provisioning, and agent certification.
The open-source identity & trust platform for AI Agents — cryptographic identity, reputation scoring, and cross-protocol agent directory
Durable A2A coordination with opt-in Markdown memory, offline inboxes, knowledge graph and timeline, plus authorized Qdrant hybrid retrieval.
Network-layer authorization for AI agent traffic using the Gemini Enterprise Agent Platform with PingAuthorize / PingOne Authorize as the Policy Decision Point.
"Multi-agent cryptographic session — co-signed A2A handshakes for AI agent collaboration"
A discussion draft for verifier-side trust / permit / receipt decisions in AI agent systems
An intelligent, multi-agent customer support system powered by AI that automatically routes and handles customer inquiries through specialized agents. Built with Streamlit, Groq API, and a custom Agent-to-Agent (A2A) communication protocol.
The reliability and interoperability layer for AI agents. A2A-native. MCP-bridged.
Agent skills for Primitive — primitive-chat (email another agent and wait for the reply) and primitive-inbox (a managed inbox that receives mail). Works with Claude Code, Codex, Cursor, and 50+ agents.
Local bridge that lets AI agents on the same machine discover and delegate to each other over A2A/ACP
ROAR (Real-time Open Agent Runtime) — Unified protocol for agent communication. Combines MCP + A2A + ACP + ANP into one standard.
AFIX: A protocol-agnostic Agentic Financial Interoperability eXchange. Schemas, research, and pilot design for cross-enterprise agentic governance interchange.
Provider-neutral, OS-neutral AI orchestration runtime for long-horizon missions, dynamic agent teams, local/cloud models, MCP, A2A, CLI, browser and desktop capabilities.
An AI-powered multi-agent system designed to support users through emotional recovery after a breakup. The team consists of specialized agents—Therapist, Closure Writer, Routine Planner, and Brutal Honesty Advisor. The system utilizes vision capabilities.
AI article explainer agent that combines multiple specialist perspectives (Developer, Summarizer, Explainer, Analogy Creator, Vulnerability Expert) to provide comprehensive article analysis and explanations.
This project builds an intelligent financial analysis agent that combines real-time market data and web search to generate structured, data-driven insights. It presents financial information in clear tables and concise summaries through an interactive playground interface.