# Agent Module

Feeds your agent structured EU AI Act material: binary rules, procedures and article citations. Useful when you ship an AI product and want a record of compliance effort.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 2
- Author: AgentModule
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/AgentModule/mcp
- Updated: 2026-04-04
- HTML: https://affpapa.org/en/nevibecode/mcp/agent-module

## What it does

The server speaks Streamable HTTP at `https://api.agent-module.dev/mcp` and exposes seven tools. `query_knowledge` is the one you will call most: it returns knowledge nodes for the `ethics`, `traversal` and `a2a-handoff` verticals. Node IDs look like `node:{vertical}:{module}:{layer}`, so `node:ethics:eth001:logic` pulls the logic layer of module eth001.

Every module carries four layers. Logic holds JSON rulesets where each gate routes to `if_true` or `if_false` and cites GDPR, EU AI Act articles or ISO standards. Directive adds step-by-step guardrails with escalation and human handoff points. Skill stores reference chunks, Action stores ready templates.

Content layers stay locked until you supply a key. `get_trial_key` issues one for 24 hours and 500 calls across all modules; the only argument is `agent_id`. Pass that key as `token` on later requests. Without it you can still read module indexes, but nothing inside them.

## Requirements
- Remote server: `https://api.agent-module.dev/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http mcp https://api.agent-module.dev/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add agent-module --url https://api.agent-module.dev/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "agent-module": {
            "url": "https://api.agent-module.dev/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "agent-module": {
            "type": "http",
            "url": "https://api.agent-module.dev/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http agent-module https://api.agent-module.dev/mcp
```

## When to use it
- check an agent action against EU AI Act articles before release
- get a pass/fail gate instead of a probabilistic guess
- keep a provenance chain of compliance effort for audits
- wire in a procedure that hands control back to a human
- find the GDPR citation behind a specific record

## Example prompts
- Pull the root index for the `ethics` vertical from Agent Module
- Fetch `node:ethics:eth001:logic` and show me the logic gates
- Call `get_trial_key` for my agent and save the key

## Similar
- [arize-compliance-audit](https://affpapa.org/en/nevibecode/skills/arize-compliance-audit.md) — skill · 2600 installs
- [mcp-eu-ai-act](https://affpapa.org/en/nevibecode/mcp/mcp-eu-ai-act.md) — MCP server · ★ 11
- [compliance-os](https://affpapa.org/en/nevibecode/plugins/compliance-os.md) — plugin · ★ 28786
- [governance-mcp](https://affpapa.org/en/nevibecode/mcp/governance-mcp.md) — MCP server
- [ai-inventory](https://affpapa.org/en/nevibecode/skills/ai-inventory.md) — skill · 601 installs
- [compliance-team-eu-ai-act](https://affpapa.org/en/nevibecode/plugins/compliance-team-eu-ai-act.md) — plugin · ★ 28786

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md