google-ads-copy

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from external sources and local files that could potentially contain malicious instructions.
  • Ingestion points: Reads search term data from Google Ads (search_term_view) and business context from {data_dir}/business-context.json and {data_dir}/personas/{accountId}.json in SKILL.md.
  • Boundary markers: The instructions do not define explicit delimiters or warnings to ignore instructions embedded within the ingested search terms or business data.
  • Capability inventory: The skill has the capability to execute GAQL queries via runScript and manage ad experiments using tools like createAdVariationExperiment and createExperiment in SKILL.md.
  • Sanitization: There is no mention of filtering or sanitizing external search term data before it is incorporated into the ad copy generation process.
  • [COMMAND_EXECUTION]: The skill uses a script execution tool to query Google Ads data.
  • Evidence: SKILL.md instructs the agent to use runScript with ads.gaqlParallel to perform data analysis on various Google Ads views.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:22 AM
Security Audit — agent-trust-hub — google-ads-copy