physics-reports
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions the agent to process content from external web searches and local files, creating a vulnerability surface where malicious instructions could influence agent behavior.
- Ingestion points: The skill explicitly directs the agent to read local files (
../../resources/source-basis.mdand../../resources/official-source-map.md) and to perform live web searches for 'Physics Reports guide for authors' on the Elsevier website. - Boundary markers: There are no boundary markers or explicit instructions provided to the agent to treat this ingested content as data only or to ignore embedded instructions (e.g., delimiters or 'ignore prior instructions' warnings are absent).
- Capability inventory: The skill is designed for venue selection and framing; it does not define dangerous capabilities such as file system writing, network exfiltration commands, or arbitrary code execution tools.
- Sanitization: No sanitization, filtering, or validation steps are prescribed for the content retrieved from external sources before it is processed by the agent.
Audit Metadata