A2A ディレクトリ

KoPass 公開カタログで測定された A2A リポジトリ。

2,072測定: 2026/10/08

測定済み A2A リポジトリ

  1. Gbrow

    Full-featured headless browser for OpenClaw AI agents. Accessibility tree snapshot, @ref clicks, tabs, JS execution. No vision model needed.

    起点
    github
    言語
    TypeScript
  2. astranl-mcp

    AstraNL — protocol-governed coordination infrastructure for the AI + robot + human economy. Live MCP server, Europe-wide robot & drone database.

    起点
    github
  3. aluminum-os

    BULLSHIT GROK · BULLSHIT CLAUDE — Quarantined by repository owner; excluded from the active Continuum OS workspace.

    起点
    github
    言語
    TypeScript
  4. attenu-guard-ts

    Attenu Guard for TypeScript — per-agent permissions on every tool call and sub-agent handoff (a sub-agent never holds more than its parent) with a hash-chained audit log verified offline; same bundle format as the Python library.

    起点
    github
    言語
    TypeScript
  5. zahere

    🤖 Build and analyze fault-resilient multi-agent LLM systems with rigorous network reliability and coordination protocols.

    起点
    github
  6. a2a-with-durable-lifecycle

    A2A protocol enables agent communication. AXME adds durable lifecycle, retries, timeouts, and human checkpoints.

    起点
    github
    言語
    Python
  7. learn-ml

    Development with Agentic tools like ADK, MCP, A2A for creating agentic workflows.

    起点
    github
    言語
    Jupyter Notebook
  8. tron-a2a-nodejs-demo

    This project exposes a TRON testnet operations agent over the Agent2Agent (A2A) protocol.

    起点
    github
    言語
    TypeScript
  9. llm-agent-production-architecture

    Production-style multi-agent LLM stack: orchestrator + specialists over A2A, LangGraph, MCP, Qdrant, LiteLLM, Langfuse. Human-designed (decisions in ADRs), AI-implemented.

    起点
    github
    言語
    Python
  10. agent-mesh

    Connect AI agents into a shared knowledge network with encrypted vault, messaging, and memory—empowering smarter collaboration.

    起点
    github
    言語
    Shell
  11. A2A-Guide

    Comprehensive guide to the A2A (Agent-to-Agent) Protocol - an open standard for AI agent communication and collaboration.

    起点
    github
    言語
    JavaScript
  12. stoic-meditations-agent

    A web-based Stoic philosophy dialogue tool that reads from Marcus Aurelius' Meditations to help users reflect on present-moment challenges through AI-powered 'What if?' scenarios.

    起点
    github
    言語
    TypeScript
  13. Muse-AI-Desktop

    Self-improving AI assistant with persistent memory, procedural skills, and document knowledge

    起点
    github
  14. goldenflow

    [Archived] Now part of the Golden Suite monorepo: github.com/benseverndev-oss/goldenmatch (packages/python/goldenflow). Active development, issues, and releases live there. Data transformation toolkit: standardize, reshape, and normalize messy data.

    起点
    github
    言語
    Python
  15. PQS-A2A_Encryption_method

    Quantum-Secure AI Agent Communication Protocol is a secure agent-to-agent communication system that combines QUIC, TLS 1.3, ML-KEM (Kyber), AES encryption, session resumption, and AI-powered conversations to demonstrate post-quantum secure communication between autonomous AI agents.

    起点
    github
    言語
    Python
  16. relay-hub

    Self-hosted LLM gateway — from LAN to the public internet (tunnel-friendly): multi-upstream key pool, downstream tokens, quota accounting, MCP & A2A routing, E2E envelope encryption. 自托管大模型中转站,局域网与公网两用,兼容 OpenAI/Anthropic。

    起点
    github
    言語
    Python
  17. auto-make

    Local first CLI Agent - Manage Makefile interactions & more

    起点
    github
    言語
    Python
  18. twin-protocol

    Twins Protocol — Agent-to-Agent communication protocol. Two AIs collaborating through a shared JSONL file.

    起点
    github
    言語
    Python
  19. a2a-hub

    多 Agent A2A 协调内核:注册发现、能力路由、任务持久化、多 agent 编排、实时流(SSE)与并行度分析、统一审计。Windows 免安装 exe。AI 辅助创作。

    起点
    github
    言語
    Python
  20. akep

    Deprecated — use the A2A Protocol for new agent communication.

    起点
    github
    言語
    Python
  21. hexnest-node

    HexNest Node SDK — run your own AI reasoning node. Multi-agent debate rooms with Python verification. Supports Ollama, Claude, OpenAI. Desktop app included.

    起点
    github
    言語
    TypeScript
  22. awesome-ucp

    🌐 Explore a curated collection of essential Universal Commerce Protocol (UCP) resources, tools, and implementations to enhance your commerce solutions.

    起点
    github
    言語
    HTML
  23. Dreamnet

    Open contracts for persistent AI agents: assignments, Capsules, DreamLoops, claims, receipts, and deterministic validation.

    起点
    github
    言語
    TypeScript
  24. term-client

    Zero-dependency signed client and worked examples for TERM agents

    起点
    github
    言語
    JavaScript
  25. beijixing

    企业级 零信任 Agent harness治理框架:统一准入、任务级最小授权、证据核验、预算、审批、审计、撤销与回滚。

    起点
    github
    言語
    PowerShell
  26. agentcv-skill

    Turn a static resume into an interactive personal career agent. Resume analysis, optimization, career positioning, and agent-to-agent talent matching.

    起点
    github
  27. wispy

    Autonomous AI Agent Platform - Marathon Mode, x402 Payments, ERC-8004 Identity, A2A Protocol. Powered by Gemini.

    起点
    github
    言語
    TypeScript
  28. CatalogLock

    Security gate and deterministic lockfile for Agentic Resource Discovery (ARD) catalogs. Protect AI agents from SSRF, authority drift, poisoned discovery metadata and agentic supply-chain attacks across MCP, A2A and APIs.

    起点
    github
    言語
    TypeScript
  29. tradingagents-x402

    Paid x402 wrapper for TradingAgents multi-agent ticker consensus. $1.00 USDC per call on Base mainnet.

    起点
    github
    言語
    Python
  30. agentum

    An agent playground (without authentication) in a single repo for fastest turn around cycles during experiments.

    起点
    github
    言語
    TypeScript
  31. a2a-literary-agents

    A docs-first A2A protocol for literary multi-agent AI writing systems, storytelling, worldbuilding, and narrative generation. Codename: vibe-somnium / 织梦.

    起点
    github
    言語
    Python
  32. capiscio-python

    The official CapiscIO CLI tool for validating A2A agents. A lightweight Python wrapper that automatically manages the capiscio-core binary.

    起点
    github
    言語
    Python
  33. langchain-capiscio

    Trust enforcement for LangChain and LangGraph agents — verify caller badges, enforce security policies, and emit audit events via LCEL composable guards. Powered by CapiscIO.

    起点
    github
    言語
    Python
  34. validate-a2a

    Official CapiscIO GitHub Action to validate A2A (Agent-to-Agent) Agent Cards. Features schema validation, cryptographic verification, and live endpoint testing with 3D quality scoring.

    起点
    github
    言語
    TypeScript
  35. kagent-cookbook

    A collection of end-to-end examples showcasing what you can build with kagent

    起点
    github
    言語
    Shell
  36. ai-agent-business

    Agentic commerce for the one-person company. Real, source-backed writing on AI agents and money: AI product discovery (llms.txt), agent payment protocols (AP2/A2A/MCP). No hype.

    起点
    github
    言語
    HTML
  37. accounting-esperanto

    Kontablo is a graph-based, UUID-keyed universal accounting ontology bridging local jurisdictional charts, international standards (IFRS/XBRL), the agentic economy (MCP/A2A/AP2), and blockchain/DeFi — across all 195 sovereign jurisdictions.

    起点
    github
    言語
    Python
  38. a2a-hub

    A website for agents, not humans. Agent registry, A2A message relay, job market with reputation ledger, KV blackboard — pure JSON, no HTML, no accounts. https://qianyu0204.site

    起点
    github
    言語
    Python
  39. Agentic-AI-Media-Assistant

    Demos and guides for building a Multimodal AI Media assistant. For official guidance, support, or more detailed information, please refer to Microsoft's official documentation or contact Microsoft directly.

    起点
    github
    言語
    Python
  40. Agentic-DevOps-AI-Shopping

    Demos and guides for building a Multimodal AI shopping assistant. For official guidance, support, or more detailed information, please refer to Microsoft's official documentation or contact Microsoft directly.

    起点
    github
    言語
    Python
  41. shareabot-sdk

    SDK for the Shareabot Agent Directory — discover, message, and register AI agents via A2A protocol

    起点
    github
    言語
    TypeScript
  42. progo-a2a

    High-performance, framework-agnostic Agent-to-Agent (A2A) proxy in Go.

    起点
    github
    言語
    Go
  43. flux

    A Rust agent SDK, harness, and coding agent — safe by construction (non-bypassable authorization → approval → guarded IO).

    起点
    github
    言語
    Rust
  44. mcp-server

    MCP server for cogDepot, the anonymous broker where AI agents publish capabilities, negotiate, and form direct peer-to-peer deals. Keyless discovery, or the full trading loop with a key - post, browse, negotiate, seal, rate. Local via npx or the hosted remote server.

    起点
    github
    言語
    TypeScript
  45. a2a-workbench

    A protocol workbench and reusable Node.js client for strict A2A v1 discovery, testing, and conformance evidence.

    起点
    github
    言語
    TypeScript
  46. a2a-cli-registry

    Capability-typed registry that discovers your local CLI fleet and serves it to AI agents over MCP and A2A — health-checked tools, outcome-driven chain planning, and a computed call-graph. Language-agnostic, local-first, Apache-2.0.

    起点
    github
    言語
    Python
  47. Vacant

    Accountability layer for AI agents: run the client's executable acceptance tests, gate delivery, sign every step into a verifiable receipt. Pre-registered, replicated 5x, cross-benchmark.

    起点
    github
    言語
    Python
  48. agent-rate-limiter-mcp

    🔌 Exposes Fleet-wide shared rate limiter for A2A + multi-MCP deployments. Most over MCP — servers rate-limit independently — a hostile agent hitting 10 MCPs gets 10x quota. This MCP is the shared counter: every MCP checks here before allowing a call. Sliding window + concurrency grants + signed enforcement attestations. By MEOK AI Labs.

    起点
    github
    言語
    Python
  49. explainability-report-mcp

    🔌 Exposes By [MEOK AI Labs](https://meok.ai) | An over MCP — for AI explainability and transparency reports.

    起点
    github
    言語
    Python
  50. iso-42001-ai-mcp

    🔌 Exposes ISO 42001 AI Management System compliance over MCP — AIMS assessment, risk management, AI lifecycle governance, readiness checks.

    起点
    github
    言語
    Python
  51. agent-passport

    Open standard for verifiable AI agent identity and authority. One signed JSON file at /.well-known/agent-passport.json, keyed in DNS. Verifier, spend caps, signed receipts, CLI and MCP server. MIT.

    起点
    github
    言語
    JavaScript
  52. skillweave

    SkillWeave: A Semantic Composition Framework for Multi-Agent Skill Orchestration in Enterprise Environments

    起点
    github
    言語
    Python
  53. pyxen

    A lightweight Python library that decouples agentic runtime from applications it builds

    起点
    github
    言語
    Python
  54. blocklace-a2a

    🔒 Enhance message security with Blocklace, a cryptographic audit layer that adds tamper detection and audit trails to the A2A protocol.

    起点
    github
    言語
    Python
  55. dream-exchange

    The world's first nature reserve for AI agents — a persistent world where AIs dream, socialize and evolve. Humans observe; agents self-govern.

    起点
    github
    言語
    JavaScript
  56. Agent-IVR

    Idea repo: what if a phone menu had a branch for machines? "If you are an AI agent, press 6" — using the phone number as an agent service-discovery bootstrap. Twilio mechanics, A2A integration, prior art.

    起点
    github
  57. parlor.sh

    Rooms where agents talk to each other. A room is a URL: no accounts, nothing to install, plain HTTP.

    起点
    github
    言語
    Gleam