caldera-mcp
Connects MCP-compatible AI clients to a MITRE Caldera adversary emulation platform, enabling natural language construction of attack scenarios, agent inspection, and operation management.
README
caldera-mcp
MCP server for MITRE Caldera adversary emulation platform.
Connects any MCP-compatible AI client (Claude, Cursor, VS Code, etc.) to a running Caldera instance. Build attack scenarios from natural language, inspect connected agents, and manage operations — all through conversation.
Important: This tool connects an AI to a live adversary emulation platform. Only point it at Caldera instances you own and are authorized to operate. Always review scope before executing any operation.
Tools (17)
| Group | Tools |
|---|---|
| Health | caldera_health_check |
| Agents | caldera_list_agents, caldera_get_agent |
| Abilities | caldera_list_abilities, caldera_get_ability, caldera_create_ability, caldera_delete_ability |
| Adversaries | caldera_list_adversaries, caldera_get_adversary, caldera_create_adversary, caldera_update_adversary, caldera_delete_adversary |
| Operations | caldera_list_operations, caldera_get_operation, caldera_create_operation, caldera_set_operation_state, caldera_get_operation_results, caldera_delete_operation |
Key behaviors:
caldera_list_abilitiessupports filters:technique_id,tactic,platformcaldera_create_operationdefaults tostate="paused"— operations never start automatically- Input validation enforced on technique IDs (ATT&CK format), platforms, and executors
Requirements
- Python 3.12+ or Docker
- A running MITRE Caldera instance (v5.x)
- The Caldera red team API key
Installation
uvx — zero install (recommended)
Requires uv.
CALDERA_URL=http://my-caldera:8888 \
CALDERA_API_KEY=your-red-api-key \
uvx caldera-mcp
pip
pip install caldera-mcp
CALDERA_URL=http://my-caldera:8888 CALDERA_API_KEY=your-red-api-key caldera-mcp
Docker (SSE / server mode)
docker run --rm \
-e CALDERA_URL=http://my-caldera:8888 \
-e CALDERA_API_KEY=your-red-api-key \
-p 8081:8081 \
ghcr.io/cowboy-samurai/caldera-mcp \
--transport sse
Configuration
| Variable | Required | Default | Description |
|---|---|---|---|
CALDERA_API_KEY |
Yes | — | Caldera red team API key |
CALDERA_URL |
No | http://localhost:8888 |
Caldera server base URL |
The red team API key can be found in your Caldera config (conf/local.yml → api_key_red), or in the container logs if auto-generated.
MCP client setup
Claude Code
Add to .mcp.json in your project root:
{
"mcpServers": {
"caldera": {
"command": "uvx",
"args": ["caldera-mcp"],
"env": {
"CALDERA_URL": "http://my-caldera:8888",
"CALDERA_API_KEY": "your-red-api-key"
}
}
}
}
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"caldera": {
"command": "uvx",
"args": ["caldera-mcp"],
"env": {
"CALDERA_URL": "http://my-caldera:8888",
"CALDERA_API_KEY": "your-red-api-key"
}
}
}
}
SSE mode (remote Caldera)
If your Caldera instance is remote and you want the MCP server to run as a persistent process:
caldera-mcp --transport sse --host 127.0.0.1 --port 8081
Then configure your client to connect via SSE:
{
"mcpServers": {
"caldera": {
"type": "sse",
"url": "http://localhost:8081/sse"
}
}
}
Usage examples
Once connected, talk to your AI client naturally:
List all alive agents
Find abilities for T1059.001 on Windows
Create an adversary from these techniques: T1566.001, T1059.001, T1003.001
What operations are currently running?
Operations created through this server default to
pausedstate. Always review scope — techniques, commands, and targeted hosts — before resuming any operation.
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.