ads-microsoft
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill's primary function involves ingesting and analyzing untrusted data from Microsoft Ads account exports and ad copy. While typical for this task, the instructions do not define boundary markers to prevent the agent from being influenced by malicious instructions embedded in the data. * Evidence Chain: (1) Ingestion points: Microsoft Ads account exports and UET status data referenced in the process section. (2) Boundary markers: Absent. (3) Capability inventory: Logic for evaluation, score calculation, and generation of MICROSOFT-ADS-REPORT.md. (4) Sanitization: None mentioned for external data strings.
- [Prompt Injection] (SAFE): No evidence of system prompt overrides, role-play bypasses, or instructions to ignore safety filters.
- [No Code] (INFO): The provided skill contains only markdown instructions and metadata. No scripts, binaries, or command executions were detected.
Audit Metadata