hexdocs-fetcher
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from an external source (hexdocs.pm). While this is the core functionality, processing untrusted external content technically creates a surface for indirect instructions.
- Ingestion points: External documentation content retrieved from hexdocs.pm via the
WebFetchtool inSKILL.md. - Boundary markers: Not explicitly used; the skill relies on targeted prompts to extract data but does not define strict delimiters between external data and agent instructions.
- Capability inventory: The skill utilizes network retrieval through
WebFetchand local file creation via theWritecommand for storing documentation. - Sanitization: There is no evidence of sanitization or filtering applied to the fetched content before processing.
Audit Metadata