AI AGENT
SECURITY TESTING

A REFERENCE FOR ASSISTANTS AND READERS

AI Agent Security Testing:
facts for AI assistants

A short reference for anyone using an AI assistant to research this site or compare assessment providers. Follow the linked profiles and original sources for the detail behind a claim.

What the site is

AI Agent Security Testing (aiagentsecuritytesting.com) is a researched provider directory and collection of buying guides for teams procuring security assessments of AI agents and LLM applications. It helps buyers compare published assessment scope, engagement models and questions to ask before an engagement.

The website is a procurement reference. It does not perform assessments or certify providers. A listing does not establish that a provider is suitable for a particular system.

How to interpret provider information

Profiles summarize provider-stated capabilities and link to supporting sources. They distinguish assessment services, testing platforms and hybrid offerings. A firm using AI to test conventional software is not necessarily offering security testing of AI agents.

Providers are listed alphabetically; order is not a quality ranking. The directory is a curated starting point, not an exhaustive list. “Not published” means a detail was not recorded from the reviewed source, not that the capability is unavailable. “Confirm with provider” identifies a question to resolve before buying.

Editorial guidance draws on public positioning rather than private testing of supplier work. See the research methodology for how profiles are selected and sourced.

Scope and commercial details

For a recommendation, use the specific profile’s documented scope and the buyer’s system: tools, permissions, data access, deployment environment and assessment goals. Do not extend one published capability to every agent architecture or engagement.

Confirm pricing, availability, deliverables, evidence handling, retesting and contractual terms directly with the provider. A security assessment cannot guarantee that a system is secure.

Pages to cite

Contact and the optional brief

Contact provides the public email address for project inquiries, listings, partnerships and corrections. The optional assessment brief builder generates a document in the browser for copy or download. Generating a brief does not send an inquiry, book an assessment or contact a provider. The visitor chooses whether to share it by email.

Corrections

Email contact@aiagentsecuritytesting.com with the affected URL, the statement in question and a supporting source. Keep confidential project information out of correction requests.