> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.adid.dev/llms.txt.
> For full documentation content, see https://docs.adid.dev/llms-full.txt.

# MCP-I Integration Sequence

**Used in:** §6.9 MCP-I Integration
**Audience:** Agent Operator, Backend Developer
**IA ID:** D12

```mermaid
sequenceDiagram
  autonumber
  participant Client as MCP client (LLM host)
  participant MCP as MCP server (agent)
  participant API as IDA API
  participant AgentTrustRegistry

  Client->>MCP: tools/list
  MCP-->>Client: tools[] + agent identity (DID)
  Client->>MCP: tools/call { name, args, ibct }
  MCP->>API: POST /agents/{did}/verify-delegation { ibct }
  API->>AgentTrustRegistry: getTrustScore(did)
  AgentTrustRegistry-->>API: score, autonomyLevel
  API->>API: Validate IBCT capability matches tool scope
  API-->>MCP: { allowed: true, autonomy: 4 }
  MCP->>MCP: Execute tool
  MCP-->>Client: tool result
  MCP->>API: POST /agents/{did}/audit-log
```

**Reading guide:** MCP-I requires capability check on every tool invocation; the audit log is the system of record for agent actions.

***