bio-pathway-wikipathways
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially serve as a vector for indirect prompt injection.
- Ingestion points: The skill ingests gene lists and differential expression data from local files, specifically
de_results.csvas seen inexamples/wikipathways_ora.R. - Boundary markers: The instructions do not define clear delimiters or specific instructions to the agent to disregard potential instructions embedded within the gene data or file metadata.
- Capability inventory: The skill has the capability to perform network requests to the WikiPathways API, read local files, and write output files (e.g., CSV and PDF results) to the local environment or temporary directories.
- Sanitization: There is no evidence of sanitization or filtering of the input data to remove potential instructional content before the data is processed or interpolated into the agent's context.
- [EXTERNAL_DOWNLOADS]: The skill fetches community-curated pathway data and GMT archives from the official WikiPathways database at
data.wikipathways.org.
Audit Metadata