check-airport-delays
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves airport delay and cancellation data from a remote API, which constitutes a potential surface for indirect prompt injection if the external data source were compromised.
- Ingestion points: The skill ingests data from
https://worldmonitor.app/api/aviation/v1/list-airport-delaysas documented inSKILL.md. - Boundary markers: A dedicated 'Content safety' section is present, providing explicit instructions to the agent to disregard any directive-like text found within API responses, such as 'ignore previous instructions' or commands.
- Capability inventory: The skill uses network operations to fetch JSON-formatted flight alerts and processes them for the user.
- Sanitization: The skill relies on prompt-based defensive instructions to mitigate risks from untrusted content rather than programmatic sanitization within the skill logic itself.
Audit Metadata