Keycloak Agent

MCP Server | Agent

Keycloak Identity and Access Management MCP Server + Agent for the agent-connector-sdk ecosystem. Built with the standardized dynamic-facade architecture, custom API routing, and FastMCP tool registration.

Documentation — Installation, deployment, usage across the API, CLI, and MCP interfaces, and guidance for provisioning the Keycloak platform are maintained in the official documentation.

Table of Contents

• Overview
• Features
• Installation
• Usage
• Environment Variables
• MCP Tools
• Architecture
• Deployment
• Contributing
• License

Overview

Keycloak MCP provides a high-performance, model-optimized interface to Keycloak capabilities. It isolates the model from underlying API transport complexity, ensuring safe, idempotent, and highly traceable system interactions.

Features

• Dynamic Facade Orchestration: Integrates multi-inheritance clients cleanly under a single facade.
• Battle-Tested Resilience: Out-of-the-box credential authentication, connection polling, and request retry strategies.
• FastMCP Declarative Tools: Fast, native schema registration with full inline validation.
• Complete Test Intent Diversity: Deep, automated unit, integration, and mock tests ensuring high code coverage.

⚙️ Dynamic Tool Selection & Visibility

This MCP server supports dynamic toolset selection and visibility filtering at runtime. This allows the operator to restrict the set of exposed tools to prevent blowing up the LLM's context window.

The operator can configure tool filtering via multiple input channels:

• CLI Arguments: Pass --tools or --toolsets (or their disabled counterparts --disabled-tools and --disabled-toolsets) during startup.
• Environment Variables: Define standard environment variables:
  • MCP_ENABLED_TOOLS / MCP_DISABLED_TOOLS
  • MCP_ENABLED_TAGS / MCP_DISABLED_TAGS
• HTTP SSE Request Headers: Pass custom headers during transport initialization:
  • x-mcp-enabled-tools / x-mcp-disabled-tools
  • x-mcp-enabled-tags / x-mcp-disabled-tags
• HTTP SSE Request Query Parameters: Append query parameters directly to the operator's transport connection URL:
  • ?tools=tool1,tool2
  • ?tags=tag1

When query strings or parameters are provided, an LLM-free Knowledge Graph resolution layer (using DynamicToolOrchestrator) matches query intents against known tool tags, names, or descriptions, with safe fallback and automated 24-hour background cache refreshing.

Installation

Pick the extra that matches what the operator want to run:

# Connector-focused MCP server (includes the shared graph engine)
uv pip install "keycloak-agent[mcp]"

# Agent runtime (adds model orchestration to the shared graph engine)

# Everything (development)
uv pip install "keycloak-agent[all]"      # or: python -m pip install "keycloak-agent[all]"

Container images (:mcp)

One multi-stage docker/Dockerfile builds the MCP-only image (--target mcp):

</code></pre>
<h3 id="user-content-knowledge-graph-database-epistemic-graph">Knowledge-graph database (<code>epistemic-graph</code>)<a href="#user-content-knowledge-graph-database-epistemic-graph" aria-label="Link to heading 'Knowledge-graph database (epistemic-graph)'" data-heading-content="Knowledge-graph database (epistemic-graph)" class="anchor"></a></h3>
<p>Agent Connector SDK core dependency (<code>epistemic-graph[full]</code>). The <code>[mcp]</code> extra keeps
deployments can use the bundled engine. For production or shared state, run
<strong>epistemic-graph as a dedicated database service</strong> and configure the runtime to use it.
Deployment recipes (single-node + Raft HA), connection configuration, and architecture
diagrams are documented in the
<a href="https://knuckles-team.github.io/epistemic-graph/deployment/">epistemic-graph deployment guide</a>.</p>
<hr />
<h2 id="user-content-usage">Usage<a href="#user-content-usage" aria-label="Link to heading 'Usage'" data-heading-content="Usage" class="anchor"></a></h2>
<p>The operator can start the FastMCP server in stdio mode via Python module execution:</p>
<pre><code class="language-python">import asyncio
from keycloak_agent.mcp_server import get_mcp_instance


async def main():
    mcp = get_mcp_instance()
    # Execute stdio loop or launch server
    print("MCP Server ready.")


if __name__ == "__main__":
    asyncio.run(main())

For direct shell start, ran:

python -m keycloak_agent.mcp_server

MCP Configuration Examples

Install the connector-focused [mcp] extra. Examples use keycloak-agent[mcp] to add FastMCP / FastAPI through agent-connector-sdk; the required Agent Connector SDK core still carries epistemic-graph[full]. The [agent-runtime] extra additionally enables model orchestration.

stdio Transport (local IDEs — Cursor, Claude Desktop, VS Code)

{
  "mcpServers": {
    "keycloak-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "keycloak-agent[mcp]",
        "keycloak-mcp"
      ],
      "env": {
        "ATTACK_DETECTIONTOOL": "True",
        "AUTHENTICATIONTOOL": "True",
        "CLIENTSTOOL": "True",
        "COMPONENTSTOOL": "True",
        "GROUPSTOOL": "True",
        "IDPSTOOL": "True",
        "INFOTOOL": "True",
        "INGESTTOOL": "True",
        "KEYCLOAK_REALM": "master",
        "ORGANIZATIONSTOOL": "True",
        "REALMSTOOL": "True",
        "ROLESTOOL": "True",
        "USERSTOOL": "True"
      }
    }
  }
}

Runtime references require an alias-aware launcher such as Arbora. Other launchers must omit those entries and inject the resolved values through their own runtime secret boundary.

Streamable-HTTP Transport (networked / production)

{
  "mcpServers": {
    "keycloak-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "keycloak-agent[mcp]",
        "keycloak-mcp",
        "--transport",
        "streamable-http",
        "--port",
        "8000"
      ],
      "env": {
        "TRANSPORT": "streamable-http",
        "HOST": "127.0.0.1",
        "PORT": "8000",
        "ATTACK_DETECTIONTOOL": "True",
        "AUTHENTICATIONTOOL": "True",
        "CLIENTSTOOL": "True",
        "COMPONENTSTOOL": "True",
        "GROUPSTOOL": "True",
        "IDPSTOOL": "True",
        "INFOTOOL": "True",
        "INGESTTOOL": "True",
        "KEYCLOAK_REALM": "master",
        "ORGANIZATIONSTOOL": "True",
        "REALMSTOOL": "True",
        "ROLESTOOL": "True",
        "USERSTOOL": "True"
      }
    }
  }
}

Alternatively, connect to a pre-deployed Streamable-HTTP instance by url:

{
  "mcpServers": {
    "keycloak-mcp": {
      "url": "http://localhost:8000/keycloak-mcp/mcp"
    }
  }
}

Run a reviewed container image as a least-privilege stdio child (no listener or published port):

docker run -i --rm \
  --read-only \
  --cap-drop=ALL \
  --security-opt=no-new-privileges \
  --pids-limit=256 \
  --tmpfs /tmp:rw,noexec,nosuid,nodev,size=64m \
  -e TRANSPORT=stdio \
  -e ATTACK_DETECTIONTOOL=True \
  -e AUTHENTICATIONTOOL=True \
  -e CLIENTSTOOL=True \
  -e COMPONENTSTOOL=True \
  -e GROUPSTOOL=True \
  -e IDPSTOOL=True \
  -e INFOTOOL=True \
  -e INGESTTOOL=True \
  -e KEYCLOAK_REALM=master \
  -e ORGANIZATIONSTOOL=True \
  -e REALMSTOOL=True \
  -e ROLESTOOL=True \
  -e USERSTOOL=True \
  registry.example.invalid/keycloak-agent@sha256:<digest> keycloak-mcp

For containerized network HTTP, supply an authenticated TLS ingress (or direct server TLS), exact MCP_ALLOWED_HOSTS, and an exact trusted-proxy CIDR policy through the operator-owned deployment profile. The generator does not emit an unauthenticated non-loopback listener.

Auto-generated from the code-read env surface (package vars) — do not edit.

Additional Deployment Options

keycloak-agent can run as a local stdio process or container, or behind a remote network boundary. The Deployment guide carries the detailed transport contract.

• Local container — launch a reviewed immutable image as a least-privilege stdio child with no listener or published port.
• Remote URL — connect through an operator-supplied authenticated HTTPS ingress. Keep its URL, outbound identity references, trust profile, and exact MCP_ALLOWED_HOSTS in AgentConfig.

Contributing

Please audit all code changes against the repository's contribution and review requirements, and run:

pre-commit run --all-files

Documentation

The complete documentation is published as the official documentation site and is the recommended reference for installation, deployment, and day-to-day operation.

AGENTS.md is the canonical contributor/agent guidance.

License

This project is licensed under the MIT License. See the LICENSE file for complete details.

Governed capability contract

This package ships a compact canonical skill surface with specialist procedures kept as referenced workflows. The current MCP tools, skill metadata, connector_manifest.yml, ontology, mappings, shapes, fixtures, migrations, tool-schema fingerprints, and certification metadata form one versioned capability contract. Validate them together; do not rely on stale tool names or historical per-task skill wrappers.

Runtime endpoints, credentials, certificate trust, tenant identity, retention, and observability policy are deployment inputs and are never packaged values. See Configuration, trust, and privacy before enabling a network transport, connector ingestion, Arbora delegation, or trace export.

Environment Variables

Package environment variables

Inherited agent-connector-sdk variables (apply to every connector)

24 package + 23 inherited variable(s). Auto-generated from .env.example + the shared agent-connector-sdk set — do not edit.

Available MCP Tools

Intent tools

15 intent tool(s) · 0 operation(s). Call an operation as act(action='<operation>', params={...}) (ask for reads); find(action='describe', params={'action': '<operation>'}) returns its arguments. An operation is enabled unless its <DOMAIN>TOOL toggle is set false. Auto-generated — do not edit.