Catalogue of Tools & Metrics for Trustworthy AI

These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.

Inner Warden



Inner Warden is an open-source, autonomous security agent for Linux and macOS servers. It detects attacks, alerts operators and, when explicitly authorised, responds automatically. The tool is relevant to AI in two ways. It can use AI models to assess threats and recommend responses. It can also monitor autonomous AI agents running on a server, such as LangChain or n8n agents, and stop harmful actions. Inner Warden runs entirely on the host, with no cloud dependency.

Inner Warden collects signals from sources such as SSH, Docker, nginx, system logs and, on Linux, extended Berkeley Packet Filter (eBPF) programs inside the kernel. Nineteen detectors, mapped to MITRE ATT&CK techniques, identify threats such as brute-force attacks, port scans, privilege escalation and container escapes. Operators can optionally connect one of twelve AI providers, including local models, to receive confidence-scored recommendations. AI remains advisory unless operators enable automatic execution and set a confidence threshold.

Response skills include blocking IP addresses, suspending user privileges, pausing containers and deploying honeypots. All actions are time-limited, reversible and recorded in an append-only audit trail with the reasoning behind each decision. The tool starts in observe-only mode, and operators can approve or deny actions through Telegram or a local dashboard. For AI agents, an application programming interface (API) lets agents check whether a command or IP address is safe before acting. Risky commands, such as download-and-execute pipelines or reverse shells, are flagged. The code is written in Rust and released under the MIT licence.

Auto-discovered on 2026-07-01 by OECD Catalogue Automation

Use Cases

There is no use cases for this tool yet.

Would you like to submit a use case for this tool?

If you have used this tool, we would love to know more about your experience.

Add use case
Partnership on AI

Disclaimer: The tools and metrics featured herein are solely those of the originating authors and are not vetted or endorsed by the OECD or its member countries. The Organisation cannot be held responsible for possible issues resulting from the posting of links to third parties' tools and metrics on this catalogue. More on the methodology can be found at https://oecd.ai/catalogue/faq.