Mindgard AI Security Testing and Red Teaming Platform

Mindgard is an AI security company founded in 2022 on more than a decade of AI security research by Dr Peter Garraghan at Lancaster University. It has a headquarters in Boston and an office in London, and in 2025 expanded its leadership with CEO James Brear and offensive security researchers Aaron Portnoy and Rich Smith.

The platform puts AI red teaming at its core. Recon performs attacker-style discovery of models, agents and tools; continuous red teaming chains adaptive attacks against RAG apps, tool-using agents and MCP or A2A servers, drawing on a knowledge base built from more than 150 publicly disclosed AI vulnerabilities; assessment shows defensive gaps; and remediation guidance, retesting and runtime threat response close the loop, with continuous assurance tracking policy compliance. It plugs into APIs, CI/CD pipelines and Burp Suite, and the team has disclosed flaws in products from Google, OpenAI and xAI.

• Recon: attacker-style discovery of models, agents and connected tools.
• Continuous red teaming: adaptive attack chaining against RAG apps, agents and MCP or A2A servers.
• Vulnerability knowledge base: built from more than 150 publicly disclosed AI vulnerabilities.
• Assessment and remediation: shows defensive gaps, gives fix guidance and retests.
• Runtime defence: AI threat response in production.
• Integrations: APIs, CI/CD pipelines and Burp Suite.

Vietnamese banks can verify that guardrails on customer chatbots resist prompt injection before each release. Security service providers can add AI red teaming to penetration testing offerings using the platform. Software companies can run AI security tests in CI/CD for agent features.

Security teams test whether guardrails and AI firewalls actually stop real attacks instead of trusting vendor claims. Continuous, automated red teaming keeps pace with new models and agent features without hiring a large offensive team. Retesting verifies that fixes work. CI/CD and Burp Suite integration fits existing application security workflows.

Guardrails and AI firewalls are spreading quickly, but new research keeps showing ways to bypass them, so buyers want independent offensive testing. Vietnamese banks, fintechs and software firms connecting agents to internal systems need the same continuous AI red teaming as global peers.

Founded in 2022 as a spin-out of Lancaster University research led by Dr Peter Garraghan; the UK entity is Mindgard Limited, registered in England. Headquarters listed in Boston with a UK office in London. Raised a seed round in 2024 and secured Fortune 500 design partners in 2026. SOC 2 Type 2 compliant.