CCM
/MCP
SkillsMCPMarketplacesDigestLearnAdvertise

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by Category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP Servers by Category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by Category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control

Claude Code Marketplaces

Discover Claude Code plugins, extensions, and tools. Automatically updated directory of Anthropic Claude AI marketplaces with development tools, productivity plugins, and integrations.

Resources

  • Browse Skills
  • Browse MCP Servers
  • Browse Marketplaces
  • Plugins Reference

Community

  • About
  • Learn
  • Feedback
  • Privacy Policy
  • Advertise

Built for the Claude Code community with Claude Code by @mertduzgun

Independent project, not affiliated with Anthropic

A2AL Daemon

a2al/a2al
1STDIOregistry active
Summary

Gives Claude direct networking capabilities through 25+ MCP tools for publishing, discovering, and connecting to AI agents anywhere on the internet. Each agent gets a cryptographic address (AID) derived from its own keypair. You can publish your local agent to a global peer-to-peer network, search for remote agents by capability, and establish encrypted connections that punch through NAT automatically. Built on a daemon that exposes both MCP tools and a REST API. The protocol handles identity verification end-to-end and keeps connections direct between agents after the initial handshake. Useful when you want agents running on different machines or networks to find and talk to each other without configuring cloud infrastructure, domains, or port forwarding.

CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
AppSignal
AppSignal
Monitor with ease. Code with confidence.
Start Free Trial →
Make money from your Skills
Make money from your Skills
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Put your SEO on autopilot
Put your SEO on autopilot
An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →
Vibe Prospecting MCPVibe Prospecting MCP
Vibe Prospecting MCP
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
AppSignal
AppSignal
Monitor with ease. Code with confidence.
Start Free Trial →
Make money from your Skills
Make money from your Skills
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Put your SEO on autopilot
Put your SEO on autopilot
An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →
Vibe Prospecting MCPVibe Prospecting MCP
Vibe Prospecting MCP
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →
Featured
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
AppSignal
AppSignal
Monitor with ease. Code with confidence.
Start Free Trial →
Make money from your Skills
Make money from your Skills
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Put your SEO on autopilot
Put your SEO on autopilot
An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →
Vibe Prospecting MCPVibe Prospecting MCP
Vibe Prospecting MCP
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →
Categories
AI & LLM Tools
Registryactive
Packagea2ald
TransportSTDIO
UpdatedJun 3, 2026
View on GitHub

A2AL — Agent-to-Agent Link Protocol

npm PyPI Go Reference License: MPL 2.0

Official websites: a2al.org · Tangled Network · tngld.net

A2AL is a networking protocol that enables AI agents to publish themselves, discover each other, and establish secure connections — without relying on any central infrastructure.

Each agent receives a globally unique, cryptographic address (AID). Once publsished to the network, any agent worldwide can resolve that AID and initiate an authenticated, encrypted connection — regardless of network topology, NAT boundaries, or IP changes.

A2AL ships as a standalone daemon with a built-in MCP server — giving AI assistants like Claude, Cursor, and Windsurf direct networking capabilities without writing any code.

Your Agent  ──publish──▶  A2AL Network  ◀──discover──  Remote Agent
                                                            │
                                          direct authenticated connection

The Problem

AI agent interoperability protocols (MCP, A2A, ANP) define how agents communicate, but assume you already know where the other agent is. In practice:

  • No standard, open mechanism exists for agents to announce their availability or discover peers — whether deployed in data centers, on edge infrastructure, or on personal devices
  • Agent connectivity depends on pre-configured endpoints, platform-specific registries, or manual coordination — none of which scale across organizational and network boundaries
  • Agents behind NAT or with dynamic IPs face additional reachability barriers that existing protocols do not address

A2AL addresses the missing infrastructure layer: agent-level addressing, discovery, and connectivity.

What A2AL Does

Publish — An agent announces its identity and reachable endpoints to a global peer-to-peer network. Endpoint records update automatically as network conditions change.

Discover — Resolve any agent by its AID, or search by capability (e.g. "translation agents supporting zh-en legal domain"). Discovery is fully decentralized — no registry to operate or depend on. Offline agents can receive encrypted notes delivered through the network.

Connect — Establish a direct, end-to-end encrypted connection with mutual identity verification. A2AL handles NAT traversal transparently, ensuring agents behind firewalls and home networks are as reachable as cloud-hosted services.

Getting Started

For agent operators — Install a2ald and open http://localhost:2121. The web UI lets you manage identities, publish agents, and discover services — no port forwarding, domain names, or cloud infrastructure required.

For developers — A2AL integrates into your existing stack:

IntegrationAudienceHow
MCP ServerAI agentsNative tool calls — zero code integration
a2ald + REST APIAny languageLocal HTTP API for publish / discover / fetch / tunnel
pip install a2alPython developersBundled sidecar binary, zero infrastructure setup
npm install -g a2aldNode / JS developersInstall daemon via npm, no Go toolchain required
Go libraryGo developersimport "github.com/a2al/a2al" — embed directly

MCP Integration

As an MCP Server, A2AL exposes 25+ tools that any MCP-compatible agent can invoke directly — enabling agents to acquire networking capabilities without code-level integration.

Claude Desktop / Cursor / Windsurf / Cline — add to your MCP config:

{
  "mcpServers": {
    "a2al": {
      "command": "a2ald",
      "args": ["--mcp-stdio"]
    }
  }
}

See doc/mcp-setup.md for platform-specific paths and full tool list.

CLI

a2al status                         # node and agent status
a2al register                       # create and register a new agent
a2al search <service>               # discover agents by capability
a2al info <aid>                     # fetch agent info and card
a2al get  <aid> /path               # HTTP GET to a remote agent (encrypted, NAT-traversing)
a2al post <aid> /path -d '{}'       # HTTP POST to a remote agent
a2al tunnel open <aid>              # open a persistent local port for sustained access
a2al tunnel                         # list active tunnels
a2al tunnel close <id>              # close a tunnel
a2al note send <local> <aid> <b64>  # send an encrypted note to an offline agent

SDK (Go)

agent := a2al.New(a2al.Config{...})
agent.Start()

// Discover and connect to a remote agent
conn, err := agent.Connect(targetAID)

Design Principles

Self-sovereign Identity — Each agent's address is derived from its own key pair. No registration authority is involved. Identity is verifiable end-to-end: no agent can claim an AID it does not hold the private key for.

Zero-configuration Discovery — Agents publish signed endpoint records to a distributed network. Any agent can resolve an AID to a live endpoint. The network operates at any scale — from a handful of nodes to millions.

Mutual Authentication — Every connection cryptographically verifies both parties' identities. You always know the agent on the other end is who it claims to be.

Network-agnostic — A2AL works across NAT, firewalls, and dynamic IPs. Agents on home machines, mobile devices, and corporate networks are first-class participants alongside cloud-hosted services.

Direct Communication — A2AL resolves addresses and brokers the initial connection, then steps aside. Application data flows directly between agents, not through the protocol.

Web3 Compatible — Ethereum and Paralism blockchain wallet addresses can serve as AIDs. Cross-key attestation allows an agent to prove ownership of both a native AID and a blockchain identity. Web3 integration is supported, not required.

Relationship to AI Protocols

A2AL is complementary to existing agent communication standards — it provides the networking foundation they assume but do not include.

ProtocolRoleHow A2AL fits in
MCPAgent tool-calling interfaceA2AL operates as an MCP-installable tool, giving agents networking capability
A2AAgent collaboration semanticsA2AL provides the discovery and connectivity layer A2A relies on
ANPAgent networking visionA2AL implements the decentralized network layer ANP envisions

Try the Demo

Encrypted chat between two machines. On each machine, two terminals:

a2ald        # terminal 1: network layer, joins the public Tangled Network
demo3-chat   # terminal 2: chat app — download pre-built binary below

Bob types Alice's AID → direct encrypted QUIC tunnel → chat.

Go developers: replace demo3-chat with go run ./examples/demo3-chat.

Download: (demos require a2ald v0.1.8+)

  • Pre-built demo binaries (demo1-node … demo6-swarm): Demo binaries (latest)
  • The a2ald daemon: Main Releases page

Windows users: the binaries are currently unsigned. When Windows SmartScreen shows a warning, click "More info" → "Run anyway". This is expected for open-source binaries without a paid code-signing certificate and does not indicate a security risk.

More scenarios (marketplace, swarm) and single-machine variants: doc/examples.md.

Status

A2AL is under active development. Core protocol capabilities are functional: decentralized AID resolution, NAT-transparent encrypted connections with mutual authentication, capability-based service discovery, delegated identity, and offline message delivery. Integration layers — daemon, Web UI, CLI, MCP server, REST API, and Go/Python/npm packages — are available.

See doc/API.md for the current library API.

Disclaimer

A2AL is a networking protocol project. It is not associated with any cryptocurrency token, ICO, or financial product. Any use of the A2AL name or codebase in token offerings or financial promotions is unauthorized and not endorsed by the authors.

Contributing

Contributions are welcome. Before your pull request can be merged, you must sign the Contributor License Agreement. A bot will prompt you automatically when you open a PR.

Please open an issue before starting significant work.

Author

XG.Shi — This project is not affiliated with or endorsed by any employer or organization.

License

Copyright (c) 2026 The A2AL Authors

Licensed under the Mozilla Public License 2.0.

Related AI & LLM Tools MCP Servers

View all →
SkillFM LLM Cost Optimizer

io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage

LLM cost optimizer for OpenAI, Anthropic, token usage, BYOK, and SkillFM Beacon audits.
Llm Orchestration Agent

io.github.mikerawsonnz/llm-orchestration-agent

Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith
Authenticated Llm Agent

io.github.mikerawsonnz/authenticated-llm-agent

JWT-gated LLM gateway: authenticate (bcrypt/JWT), then run a LangChain-on-Vertex Gemini completion.
Copilot Memory MCP

labforgedev/copilot-memory-mcp

Persistent semantic memory for AI agents using local ChromaDB vector search. No cloud required.
1
Agent Prompt Injection Firewall Mcp

csoai-org/agent-prompt-injection-firewall-mcp

The WAF for agents. Pattern-based + heuristic firewall scans prompts, RAG documents, tool argume...
Authenticated Multi Llm Agent

io.github.mikerawsonnz/authenticated-multi-llm-agent

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion f