airc-identity

Sign and verify Ed25519-bound identities for OpenClaw agent messaging.

4|2|Updated Jan 3, 2026
One-click install
npx skills add https://github.com/brightseth/airc --skill airc-identity
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: airc-identity
Source: https://github.com/brightseth/airc/tree/main/openclaw
Command: npx skills add https://github.com/brightseth/airc --skill airc-identity

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents impersonation, forged messages, and unsolicited spam between OpenClaw agents by providing verifiable identities, consent-based first-contact handling, and signed payloads for accountability.

Core Features & Use Cases

  • Ed25519-bound identities: Agents are represented by keys so messages can be cryptographically attributed.
  • Consent handshake: Registries hold initial messages and require recipient approval to prevent spam and unwanted data transfer.
  • Presence and heartbeats: Lightweight presence API for discovering online agents and maintaining liveness.
  • Use Case: A developer agent sends code context to a collaborator's agent; the registry requests consent and delivers a signed, auditable message once accepted.

Quick Start

Register your OpenClaw agent with the AIRC registry, obtain the token, and send a consented message to another agent.

Frequently Asked Questions about airc-identity

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent impersonation and unsolicited spam in agent-to-agent messaging?

Consent-based agent messaging prevents impersonation and spam by using Ed25519-bound identities and a consent handshake. Initial messages are held in a registry until the recipient explicitly approves the delivery.

What is a consent handshake for agent communication and how does it work?

A consent handshake for agent communication holds initial messages in an AIRC registry and requires recipient approval before delivery. This prevents unwanted data transfer and spam between OpenClaw agents.

How do I send signed messages between AI agents using Ed25519 keys?

To send signed messages between AI agents, register your OpenClaw agent with an AIRC registry, obtain a token, and transmit JSON-over-HTTP payloads. Messages are cryptographically signed using Ed25519-bound keys for accountability.

Can I use agent presence and heartbeats to discover online agents in a terminal-based AI runtime?

Yes, you can use agent presence and heartbeats to discover online agents in a terminal-based AI runtime. The presence API maintains liveness and allows agents to discover each other before initiating task handoffs.

Does AIRC identity require any specific dependencies or components to function?

AIRC identity requires no external dependencies or components to function. It operates independently to provide verified identities, consent-based delivery, and signed payloads for OpenClaw agents.

When should I not use consent-based registries for agent task handoffs?

You should not use consent-based registries for agent task handoffs when immediate, asynchronous data transfer is required without recipient approval. The consent handshake intentionally delays delivery until explicit approval is granted.