mysql-docs
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from external websites (dev.mysql.com).
- Ingestion points: The workflow instructs the agent to search and read pages from the live MySQL Reference Manual at
dev.mysql.com/doc/refman/. - Boundary markers: The skill does not explicitly define markers to separate retrieved documentation content from system instructions, though it requires linking claims to specific pages.
- Capability inventory: The skill is limited to reading documentation and answering questions; it has no capabilities for file system modification, command execution, or network exfiltration.
- Sanitization: No explicit sanitization of the retrieved HTML/Markdown content is mentioned.
- [SAFE]: All external references target
dev.mysql.com, which is the official and trusted domain for Oracle's MySQL documentation. This alignment with the skill's primary purpose and use of well-known, official sources presents no significant security risk.
Audit Metadata