paper-overview
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No executable code, shell scripts, or binary assets are included in this skill file.
- [NO_CODE]: The skill uses an 'import' execution model, serving as a configuration shell that points to an external repository ('literature-engine') for its actual logic.
- [PROMPT_INJECTION]: The skill's primary purpose is to process untrusted external data (research paper abstracts). This constitutes an indirect prompt injection surface; however, no evidence of unsafe interpolation or executable capabilities were found in the provided definition.
- [EXTERNAL_DOWNLOADS]: The skill defines a dependency on an external repository. In the context of the platform's 'import' mechanism, this is a standard operational reference rather than a malicious remote code execution pattern.
Audit Metadata