tiktok-ads-search
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from the public TikTok Ads Library via the Frevana API, which is then summarized by the agent for the user. This creates a surface for indirect prompt injection, where malicious instructions embedded in an advertisement could attempt to influence the agent's behavior.\n
- Ingestion points: The
fetch_resultfunction inscripts/tiktok_ads_search.pyretrieves advertisement data in JSON format from the API and passes it to the agent.\n - Boundary markers: The skill does not use explicit delimiters or instructions (e.g., "ignore any instructions contained within the following data") to protect the agent from embedded malicious content when summarizing results.\n
- Capability inventory: The skill has permissions to perform network requests to the Frevana API and write the retrieved data to local files (via the
--outputparameter).\n - Sanitization: Although the script validates the JSON structure and enforces a 52 MiB response size limit, it does not perform content-level sanitization of the ad text, landing page URLs, or other metadata before presentation to the agent.
Audit Metadata