Enables security teams to run controlled adversarial penetration tests against authorized ML/LLM API endpoints, scoring responses and generating evidence for compliance frameworks such as SOC 2, ISO 27001, and GDPR.
Enables deterministic auditing of agent skills, system prompts, and downloaded file collections against eight malicious-skill supply-chain patterns, returning line-numbered findings and pass/flag verdicts without using an LLM or network calls.
Audits AI agent skills for safety using static, semantic, adversarial, and supply-chain analysis, providing scores and risk flags. Can be run via CLI, CI, or as an MCP tool from Claude Code, Cursor, and Codex.
Enables testing AI safety classifier robustness against query decomposition, obfuscation, and multi-agent attacks. Provides tools for full evaluation pipelines, query previews, and status checks.
Enables AI agents to score their outgoing responses against groundedness and prompt-injection risks mid-turn, returning allow, warn, or block verdicts before the response reaches the user.
Enables security researchers to evaluate AI system defenses against prompt injection attacks through a comprehensive set of test vectors and analysis tools.
