sd-parse-results
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection by processing untrusted data from an external website.
- Ingestion points: The JavaScript block in
SKILL.mdextracts text content (titles, authors, journal names) directly from the ScienceDirect DOM. - Boundary markers: There are no explicit delimiters or instructions to the agent to ignore potential commands embedded in the scraped text.
- Capability inventory: The skill uses
evaluate_scriptto read the page content and returns the data to be consumed by other skills. - Sanitization: The skill does not perform any validation or escaping of the extracted strings before they enter the agent's context.
Audit Metadata