train
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from user-provided files and URLs to extract rules, standards, and glossary terms.
- Ingestion points: The skill reads data from file references (e.g.,
@docs/style-guide.md) or external URLs using thewebfetchtool. - Boundary markers: There are no explicit instructions or delimiters defined to separate the untrusted document content from the agent's primary instructions.
- Capability inventory: The skill utilizes
webfetchfor network access andtraining_saveto persist extracted information in the agent's memory. - Sanitization: No sanitization or validation logic is specified to prevent malicious instructions hidden within documents from being interpreted as valid training rules.
Audit Metadata