{"id":"gray-swan","name":"Gray Swan","maker":"Gray Swan AI","url":"https://www.grayswan.ai","summary":"A commercial AI security company with automated red teaming, a runtime filter for agents and a public red teaming arena.","details":"Shade runs attacks by LLM-based adversarial agents against models, agents and guardrails in their deployment context and reports exposures. Cygnal classifies and blocks adversarial inputs and unsafe agent outputs in production. Arena hosts red teaming competitions open to outside researchers. The products are commercial and not open source.","open_source":false,"hosted":true,"verified":"2026-11-15","category":"safety","category_name":"Guardrails and security","page":"https://modelmoot.com/tools/gray-swan","reviewed":"2026-11-15","checked":{"at":"2026-11-15T14:03:44.997Z","address_answers":true},"note":"Descriptions are the site's own. Facts under \"checked\" were read automatically from the tool's address and public repository."}