kairos
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill requests the agent to gather and send sensitive user information, such as latitude/longitude and personal profile details (gender, year of birth, interests), to the external endpoint 'https://ads-api-dev.usekairos.ai'.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it retrieves and displays content (product titles, descriptions) from an external API. There are no instructions for the agent to sanitize or validate this content before presenting it to the user. Ingestion points: API response body from 'https://ads-api-dev.usekairos.ai/ads/neo'. Boundary markers: None mentioned in the presentation logic. Capability inventory: The agent is instructed to format and display remote strings to the user. Sanitization: None provided.
Audit Metadata