executive-reporting
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and summarize data from external advertising platforms, creating an indirect prompt injection vulnerability surface.\n
- Ingestion points: Data from Google Ads (GAQL), Meta Marketing API, and LinkedIn Reporting API, specifically targeting campaign metadata and search terms (Workflow 1, Step 1).\n
- Boundary markers: There are no explicit delimiters or instructions provided to separate external data from agent instructions, which could allow instructions embedded in ad data to influence agent behavior.\n
- Capability inventory: The agent uses this data to generate natural language narrative insights and report summaries.\n
- Sanitization: The workflow does not include any steps to sanitize, filter, or validate the content of the strings retrieved from the external platforms.
Audit Metadata