competitor-ad-intelligence
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and analyzes untrusted data from external sources (competitor ad copy and landing page content). This content could contain instructions designed to manipulate the agent's summary or strategic analysis.
- Ingestion points: External content retrieved from the Meta Ad Library, Google Ads Transparency Center, and competitor landing pages via
web_searchandfetch_webpage(SKILL.md). - Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the fetched data.
- Capability inventory: Capabilities are limited to data retrieval using
web_search,fetch_webpage, andcurl(SKILL.md). - Sanitization: No sanitization or filtering of the retrieved content is specified before analysis.
- [EXTERNAL_DOWNLOADS]: The skill downloads data from various remote websites, including social media platforms and third-party competitor domains, to perform its analysis.
- [COMMAND_EXECUTION]: The skill directs the agent to use the
curlcommand-line tool as a fallback mechanism for retrieving landing page content from external URLs.
Audit Metadata