MITRE Caldera
caldera.apache.org
- Free plan
- macOS
- Linux
- Web
- API
- Self-hosted
What MITRE Caldera is
MITRE Caldera, also known as Apache Caldera, is an open-source framework for automating adversary emulation and routine cybersecurity assessments. Built on the MITRE ATT&CK framework, it uses adversary profiles to exercise defenses and train blue teams. Listed uses include autonomous emulation, testing detection and response platforms, manual red-team work, and red-versus-blue research. Its core has an asynchronous command-and-control server, REST API, and web interface, while plugins add capabilities such as Atomic Red Team techniques, operational technology support, ATT&CK visualizations, incident response, reporting, and training. The default Sandcat agent is written in Go and supports cross-platform deployment, beaconing, instruction execution, payload downloads, and file uploads. Caldera can be installed from source or through Docker. The project lists Linux or macOS and Python 3.10 or later with pip, and recommends at least 8 GB RAM and 2 CPUs for the core framework. It recommends a secure network or environment because the web interface is not hardened or thoroughly penetration-tested and has only basic security features. The software is free under Apache License 2.0 and intended for on-premises deployment.
Who it suits
It is suited to security teams and researchers conducting adversary emulation, red-team engagements, or detection and response testing. Users should be able to deploy and secure an on-premises framework.
Good to know, and what to watch
- Built on the MITRE ATT&CK framework
- Supports adversary emulation and blue-team training
- Plugins cover OT, incident response, and reporting
- Offers source and Docker installation
- Free under Apache License 2.0
- Web interface has only basic security features
- Project says the interface is not hardened
- Builder plugin does not work in Docker
- Container data is ephemeral by default
Plans
Every plan and priceApache Caldera
Open source
The verdict
Caldera brings an extensible, ATT&CK-based framework to on-premises security exercises. Its own guidance calls for a secure deployment environment, and Docker users should note the Builder plugin limitation and ephemeral container data.
What the maker publishes
- Agent
- The default Sandcat agent is written in Go and supports cross-platform deployment, beaconing, instruction execution, payload downloads, and file uploads. source
- Architecture
- The core includes an asynchronous command-and-control server, a REST API, and a web interface, with plugins that add functionality. source
- ATT&CK
- Caldera is built on the MITRE ATT&CK framework and can use adversary profiles to test defenses and train blue teams. source
- Authentication
- Caldera supports login through its internal user mapping or LDAP, and its configuration can specify a custom login handler. source
- Deployment
- The project documents installation from source and Docker deployment, including full and slim container variants. source
- Host requirements
- The repository lists Linux or macOS, Python 3.10 or later with pip, and recommends at least 8 GB RAM and 2 CPUs for the core framework. source
- Integrations
- The Atomic plugin imports tests from the open-source Red Canary Atomic tests repository. source
- Notable limits
- The repository states that the Builder plugin does not work in Docker and that Caldera container data is ephemeral by default. source
- OT support
- Caldera for OT plugins add support for common industrial protocols through collections of protocol-specific abilities. source
- Plugins
- Team-maintained plugins include Atomic Red Team TTPs, Caldera for OT, ATT&CK visualizations, incident response, reporting, and training. source
- Purpose
- Apache Caldera automates adversary emulation and routine cybersecurity assessments. source
- Security
- The project recommends running Caldera in a secure environment or network and says its web interface is not hardened or thoroughly penetration-tested and has only basic security features. source
- Support and learning
- The project links to documentation, training, use cases, tutorial videos, a blog, Discord, and GitHub discussions. source
- Use cases
- The project describes autonomous adversary emulation, detection and response platform testing, manual red-team engagements, and red-versus-blue research as use cases. source
Checked against the maker's pages on 5 Oct 2026.