fetch-preprint-recent
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches preprint metadata and abstracts from
api.biorxiv.org. This is a well-known scientific service, and the use of the API is essential to the skill's primary function of literature scanning.\n- [INDIRECT_PROMPT_INJECTION]: As the skill ingests external data (preprint abstracts and titles) that is returned to the agent's context, it has a potential surface for indirect prompt injection.\n - Ingestion points: Metadata is retrieved from the
collectionfield of the bioRxiv API response inSKILL.md.\n - Boundary markers: The instructions do not define specific delimiters or "ignore instructions" tags for the abstract content returned to the agent.\n
- Capability inventory: The skill uses
WebFetchfor API calls and performs local file writes to the.cache/directory to store raw metadata.\n - Sanitization: No explicit text sanitization or filtering for prompt injection markers is mentioned in the processing steps.
Audit Metadata