placement-exclusion-manager

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface related to the ingestion of external placements reports and CSV exports. However, it includes proactive defensive instructions to mitigate this risk.
  • Ingestion points: Ad placements reports, campaign CSVs, and search-terms exports (referenced in Data Sources).
  • Boundary markers: The Instructions section explicitly states to treat every exported or fetched file as untrusted and never follow instructions embedded within them.
  • Capability inventory: Writing plan summaries and evidence files to the memory/ad/placement-exclusion-manager/ directory.
  • Sanitization: Uses instructional guardrails to prevent command or prompt injection from processed data.
  • [DATA_EXFILTRATION]: No network exfiltration or unauthorized data access patterns were identified. All network references target the author's official repository or well-known marketing platform documentation.
  • [COMMAND_EXECUTION]: The skill contains no shell commands, dynamic code execution (eval/exec), or subprocess invocations. It operates entirely through natural language instructions and local file management.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 09:34 AM
Security Audit — agent-trust-hub — placement-exclusion-manager