relevance-coarse-filter
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified during the analysis of the skill's instructions and metadata.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill does not request access to sensitive file paths, environment variables, or credentials. It does not perform network operations to exfiltrate data.
- [PROMPT_INJECTION]: The instructions are strictly task-oriented, focusing on the classification of news signals. There are no attempts to override agent safety protocols or ignore prior instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from news signals (titles and excerpts). While it lacks explicit boundary markers (e.g., XML tags or delimiters) to isolate this untrusted content, the skill's capabilities are restricted to producing a structured JSON classification. This limited output format effectively mitigates the risk of an indirect prompt injection attack influencing the agent to perform unauthorized actions.
- [DYNAMIC_CONTEXT_INJECTION]: The skill does not utilize the dynamic context execution syntax.
Audit Metadata