> ## Documentation Index
> Fetch the complete documentation index at: https://docs.spellguard.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AMP Overview

> Admissible Messaging Policy for tamper-evident audit logs

# AMP: Admissible Messaging Policy

AMP provides the infrastructure for proving what happened during AI agent interactions. It generates cryptographic commitments that prove messages existed without revealing their contents.

## Components

<CardGroup cols={2}>
  <Card title="Message Structure" icon="envelope" href="/specifications/amp/messages">
    Encrypted message format for secure communication
  </Card>

  <Card title="Commitments" icon="certificate" href="/specifications/amp/commitments">
    Cryptographic proofs of message existence
  </Card>

  <Card title="Logging Backends" icon="database" href="/specifications/amp/backends">
    Pluggable storage for commitments and archives
  </Card>

  <Card title="Unilateral Mode" icon="arrow-right" href="/specifications/amp/unilateral">
    Communication with A2A-only agents
  </Card>
</CardGroup>