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

# AgentLedger: Trust & Discovery for Autonomous Agents

> AgentLedger fills the trust gap left by MCP, A2A, and agents.txt with a Manifest Registry, Trust Ledger, and Audit Chain for the autonomous agent web.

AgentLedger is trust and discovery infrastructure for the autonomous agent web. This page explains what AgentLedger is, the scale of the trust problem it addresses, and how its three components work together to fill a gap that no existing protocol covers.

## What AgentLedger does

The protocols that define agent communication today — MCP, A2A, and agents.txt — explicitly scope out trust. When an agent executes a financial transaction, books travel, or shares medical records with a third-party service, no infrastructure exists to answer the foundational question:

> **Can this service be trusted?**

AgentLedger is a three-component infrastructure layer that sits **above** existing communication protocols and **below** orchestration frameworks — filling the trust gap that no existing protocol addresses.

<CardGroup cols={3}>
  <Card title="Manifest Registry" icon="database" href="/architecture/manifest-registry">
    Universal capability ontology and semantic discovery API. DNS + App Store + NPM, built agent-first.
  </Card>

  <Card title="Trust Ledger" icon="shield-check" href="/architecture/trust-ledger">
    Blockchain-anchored attestation records. Immutable, cross-registry, cryptographically verifiable.
  </Card>

  <Card title="Audit Chain" icon="link" href="/architecture/audit-chain">
    Tamper-proof action logs enabling liability attribution for autonomous agent transactions.
  </Card>
</CardGroup>

## The scale of the problem

The autonomous agent web is fragmenting faster than any standards body can respond. The numbers below reflect the state of agent infrastructure as of Q1 2026.

| Metric                   | Reality                              |
| ------------------------ | ------------------------------------ |
| Agents registered        | 104,000+ across 15+ registries       |
| IETF discovery drafts    | 10+ competing, no converged standard |
| Known MCP threat vectors | \~40 with no unified defense         |
| Cross-registry trust     | Zero                                 |

## Why now

The standards landscape is at a critical inflection point. Domain-specific registries are forming independently — Google UCP for commerce, IAB AAMP for advertising, Huawei A2A-T for telecom. Each uses incompatible trust models. The window to establish a neutral, open interoperability layer is measured in months, not years.

<Card title="Read the full problem statement" icon="arrow-right" href="/problem">
  Understand the discovery gap, trust vacuum, and fragmentation crisis in detail.
</Card>

## Where to go next

<CardGroup cols={2}>
  <Card title="Architecture overview" icon="diagram-project" href="/architecture/overview">
    See how the three components work together as an integrated infrastructure layer.
  </Card>

  <Card title="Manifest Registry" icon="database" href="/architecture/manifest-registry">
    Learn how services register capabilities and how agents query them.
  </Card>

  <Card title="Trust Ledger" icon="shield-check" href="/architecture/trust-ledger">
    Understand how attestation records are anchored on-chain and verified.
  </Card>

  <Card title="Audit Chain" icon="link" href="/architecture/audit-chain">
    See how every agent action is captured in a tamper-proof chain of custody.
  </Card>
</CardGroup>