Agent Barn is an open-source control plane for running and governing AI agents on an organisation's own infrastructure. It was built by AAI Labs in Lithuania. Agents connect to team chat tools such as Slack and Microsoft Teams, and act on systems such as GitHub and Jira. Access roles control who manages each agent. Audit logs record conversations and tool calls, and costs are attributed to each agent. It is released under the Apache 2.0 licence.

The CREST Accreditation Standards set requirements for organisations that provide cyber security services, verified through independent assessment. They are published by CREST, an international not-for-profit body. Three components address AI. Domain 7 requires all accredited organisations to govern their use of AI, including validating AI-supported outputs. Annex B sets requirements for using AI in penetration testing, with human validation of findings. A third standard covers security testing of generative AI and large language model-enabled systems.

Mandare is an open-source accountability system for fleets of AI agents. It gives each agent a signed identity and a signed mandate setting its spending caps, permissions and approval requirements. A gateway enforces these limits outside the agent, with an offline kill switch. Every action, including refusals, is recorded in a tamper-evident ledger that an independent witness can check. Raw data never leaves the user's machine. It is released under the AGPL-3.0 and Apache 2.0 licences.

DelusionEval is a dataset for evaluating how AI chatbots behave when conversations reinforce users' delusional beliefs. It was developed by Stanford University. It contains 725 excerpts from real, anonymised conversations, each labelled with one of 18 chatbot behaviours. These include harmful behaviours, such as endorsing delusions or facilitating self-harm, and protective ones, such as discouraging violence. Researchers use it to audit chatbot safety and develop detection methods. Access is restricted to non-commercial research.

AI Compliancy is an online EU AI Act risk assessment, obligation and report tool for UK small and medium-sized businesses that use AI built into everyday software. Users check whether the Act can reach them, record the software they use, assess what they use each AI feature for, work through the obligations that follow, and produce a dated report.

Inner Warden is an open-source security agent for Linux and macOS servers. It detects attacks such as brute-force attempts and privilege escalation, and alerts operators in real time. It can use AI models to recommend responses, but AI remains advisory unless operators allow automatic action. Inner Warden can also monitor autonomous AI agents and block risky commands. All actions are reversible and recorded in an audit trail. It runs locally and is released under the MIT licence.

AI Facts is a free, open-source Decision Terrain tool with an independently developed checklist mapped to NIST AI RMF 1.0. It is not affiliated with, sponsored by, or endorsed by NIST. Teams document self-reported safeguards, evidence notes, gaps, owners, and next actions, then export assessment reports and AI Facts labels. This public-alpha tool does not verify evidence or establish certification, compliance, or safety.

Robust and Reliable Algorithmic Recourse (ROAR) is a framework for generating instance-level algorithmic recourse that is designed to remain reliable when the underlying predictive model changes. It helps practitioners evaluate and improve the robustness of recourse/decision-support explanations so that suggested actions continue to work under model or distribution shifts. Target users include researchers and developers working on fair/robust recourse systems to address robustness and accountability-related trustworthiness objectives.

Human Approval Gate is a free, platform-neutral educational kit that helps leaders, educators, operators, and small teams define what a qualified person must check before AI-assisted work can affect a real decision or action. It includes a practical guide, printable worksheet, facilitator notes, and ten synthetic test cases. Its CLEAR test holds the consequence, names the reviewer and evidence, preserves accept, revise, reject, and escalate outcomes, and records the decision and recovery path.

Microsoft Agent Governance Toolkit is an open-source framework for governing AI agents at scale. The toolkit provides identity management, policy enforcement, authorization, and audit capabilities, helping organizations control agent actions and maintain accountability across agent workflows and enterprise systems.

An open-source tool that audits the human-AI interaction layer of a decision-support AI system: how it presents its results, whether the person can correct it, and whether its alerts fire at the right moment. It scores that layer against Microsoft's HAX-18 and Google's PAIR design guidelines and returns concrete, evidence-anchored findings mapped to the EU AI Act and the NIST AI RMF.

A three-layer, enterprise-wide AI risk management and governance framework that operationalises trustworthy AI from board strategy to operational controls and organisational resilience.

The Deployer AI Risk Register (DARR) is an open catalogue of 82 AI risks and 61 security sub-risks for organisations deploying AI systems. It consolidates the MIT AI Risk Repository with ISO/IEC standards, MITRE ATLAS and the EU AI Act, and groups risks into seven families aligned with existing enterprise risk functions. Each risk has a stable identifier and mappings to external frameworks. The data is freely available in CSV and JSON under a CC BY 4.0 licence.

A practical guide aimed at empowering citizens and professionals to effectively utilize Artificial Intelligence in everyday life and at work, without requiring prior technical knowledge. It provides prompt engineering skills and promotes responsible AI use.

This handbook provides guidance on ethical and legal responsibilities associated with the use of high-risk AI systems in the EU civil security domain, focusing on use cases in border control, policing, and immigration. The aims of the handbook are to provide a structured pathway for engaging and understanding deployer responsibilities as outlined in the AI Act, focusing on high-risk AI systems, and to help establish processes for the ethical use of AI solutions.

Amazon Nova Premier is a multimodal foundation model that was evaluated under Amazon’s Frontier Model Safety Framework to assess and mitigate risks related to Chemical, Biological, Radiological, and Nuclear (CBRN) weapons proliferation, offensive cyber operations, and automated AI research and development.

Vaara is an open-source governance and evidence layer for AI agents. It checks each agent action against a defined policy before it runs, and allows, blocks or escalates it for human review. Every decision and outcome is written to a tamper-evident audit trail. Independent parties can verify these records offline without trusting the organisation. Vaara helps deployers assemble evidence for EU AI Act obligations, but does not certify compliance. It is released under the AGPL-3.0 licence.

Fuel iX is an enterprise AI platform that enables organisations to connect their infrastructure to a library of large language models and build, deploy and manage generative AI applications with centralized control and observability.

VERA-MH (Validation of Ethical and Responsible AI in Mental Health) is a comprehensive framework for evaluating AI chatbots in a mental health context.

Participatory Harm Auditing Workbenches and Methodologies (PHAWM) is a guiding framework and online tool for participatory AI auditing. It enables people without a technical AI background to audit the harms of AI applications. The methodology guides organisations through planning and completing an audit. The workbench takes auditors through four phases: understand, define, evaluate and recommend. Audit results inform decisions on whether to adopt AI. PHAWM was co-designed with end-users and affected people, and is free to use.