ad-negative-keywords
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified in the skill instructions or referenced files. All external references are to other skills within the same author's namespace.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential surface for indirect prompt injection by processing search term reports containing strings generated by anonymous internet users.
- Ingestion points: The skill ingests untrusted search term data from external reports for the purpose of analysis and classification.
- Boundary markers: The instructions do not define specific delimiters or security warnings to treat the search term data as untrusted content.
- Capability inventory: The skill generates text-based analysis, tables, and summaries. It does not utilize capabilities for file system modification, network communication, or shell command execution.
- Sanitization: No explicit sanitization or filtering of the input search terms is implemented.
Audit Metadata