google-ads-scripts

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill includes templates that use MailApp.sendEmail() and SpreadsheetApp.openByUrl() to export campaign data. These are standard features for the intended use case of alerting and reporting but represent a mechanism for moving account data to external endpoints defined by the user.
  • [INDIRECT_PROMPT_INJECTION]: The 'Search Term Mining' template in SKILL.md ingests untrusted strings from search query reports.
  • Ingestion points: AdsApp.report retrieves data from the SEARCH_QUERY_PERFORMANCE_REPORT table.
  • Boundary markers: Absent; raw query strings are processed directly from the report rows.
  • Capability inventory: The script has the capability to modify account settings via negList.addNegativeKeyword() and output data to logs.
  • Sanitization: The script uses regex tests (.test(query)) to identify specific patterns before adding them as exact-match negative keywords, which limits the impact of potentially malicious search strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 09:15 AM
Security Audit — agent-trust-hub — google-ads-scripts