agent-mesh-architecture

agent-mesh-architecture is a skill for Claude Code from latestaiagents/agent-skills. It costs 49 tokens per session (2,503 once invoked), scanned A, original, MIT.

A design guide for peer-to-peer systems in which multiple AI agents communicate and cooperate without one central controller.

In plain words
What is it for?
Use it to design agent networks, swarm-like collaboration, peer discovery, and direct message passing between agents.
Why use it?
It helps avoid a single point of failure and supports systems where agents can join, leave, or contribute different perspectives.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the agent-architect plugin — 13 skills, 1 command, 5 MCP servers shipped together

Good fit Use it to design agent networks, swarm-like collaboration, peer discovery, and direct message passing between agents.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/latestaiagents/agent-skills/agent-mesh-architecture
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add latestaiagents/agent-skills --skill agent-mesh-architecture
Clone the repo
git clone --depth 1 https://github.com/latestaiagents/agent-skills

Made for: Claude Code.

Or install agent-architect, the plugin that ships this one along with the rest of its 13 skills, 1 command, 5 MCP servers.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for agent-mesh-architecture

README.md
[![agentmods](https://agentmods.dev/badge/skills/latestaiagents/agent-skills/agent-mesh-architecture/github.svg)](https://agentmods.dev/skills/latestaiagents/agent-skills/agent-mesh-architecture)
Your own site
<a href="https://agentmods.dev/skills/latestaiagents/agent-skills/agent-mesh-architecture"><img src="https://agentmods.dev/badge/skills/latestaiagents/agent-skills/agent-mesh-architecture/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent-mesh-architecture

Your own site · 80×15
<a href="https://agentmods.dev/skills/latestaiagents/agent-skills/agent-mesh-architecture"><img src="https://agentmods.dev/badge/skills/latestaiagents/agent-skills/agent-mesh-architecture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,503 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00049 $0.02503
Opus 5 $0.00024 $0.01252
Sonnet 5 $0.00010 $0.00501
Haiku 4.5 $0.00005 $0.00250

Measured 12d ago against content hash 4a5f1e1c2880, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

agent-mesh-architecture scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/agent-architect/skills/patterns/agent-mesh-architecture/SKILL.md · 403 lines

How it starts

The opening of the file, as written. The whole thing — 403 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Mesh Architecture

Design resilient peer-to-peer agent networks where agents collaborate without central orchestration.

When to Use

  • Resilience is critical (no single point of failure)
  • Agents need to collaborate dynamically
  • Tasks benefit from emergent behavior
  • Scale varies (agents join/leave dynamically)
  • Different perspectives improve outcomes

Architecture Overview

    ┌─────────┐     ┌─────────┐
    │ Agent A │◄───►│ Agent B │
    └────┬────┘     └────┬────┘
         │               │
         │   ┌───────┐   │
         └──►│ Agent │◄──┘
             │   C   │
         ┌──►│       │◄──┐
         │   └───────┘   │
    ┌────┴────┐     ┌────┴────┐
    │ Agent D │◄───►│ Agent E │
    └─────────┘     └─────────┘

Core Components

1. Agent Node

Each agent is an independent node with communication capabilities.

interface AgentNode {
  id: string;
  capabilities: string[];
  state: AgentState;

  // Communication
  broadcast(message: Message): Promise<void>;
  sendTo(agentId: string, message: Message): Promise<void>;
  onMessage(handler: MessageHandler): void;

  // Discovery
  discoverPeers(): Promise<AgentNode[]>;
  announceCapability(capability: string): void;

  // Task handling
  canHandle(task: Task): boolean;
  handle(task: Task): Promise<Result>;
}

interface Message {
  type: 'task_request' | 'task_result' | 'collaboration_invite' |
        'capability_query' | 'heartbeat' | 'consensus_vote';
  from: string;
  to: string | 'broadcast';
  payload: unknown;
  timestamp: Date;
  correlationId: string;
}

2. Message Bus / Communication Layer

interface MessageBus {
  publish(topic: string, message: Message): Promise<void>;
  subscribe(topic: string, handler: MessageHandler): Subscription;
  request(agentId: string, message: Message): Promise<Message>;
}

// In-memory implementation for local agents
class LocalMessageBus implements MessageBus {
  private subscribers = new Map<string, MessageHandler[]>();

  async publish(topic: string, message: Message) {
    const handlers = this.subscribers.get(topic) || [];
    await Promise.all(handlers.map(h => h(message)));
  }

  subscribe(topic: string, handler: MessageHandler) {
    const handlers = this.subscribers.get(topic) || [];
    handlers.push(handler);
    this.subscribers.set(topic, handlers);

    return {
      unsubscribe: () => {
        const idx = handlers.indexOf(handler);
        if (idx >= 0) handlers.splice(idx, 1);
      }
    };
  }
}

Read the full file on GitHub · 403 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 403 lines · 49 tokens per session scan A 4a5f1e1c2880

Subscribe to this mod's changes

agent-mesh-architecture is a skill published in the GitHub repository latestaiagents/agent-skills (5 stars, last pushed 4mo ago), licensed MIT. It adds 49 tokens to every session and 2,503 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.