email-issue-fixer
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted email content provided by the user, creating a surface for indirect prompt injection. However, the skill lacks dangerous capabilities such as network access, file system modification, or command execution. It also includes specific instructions to identify and redact sensitive information like tokens or credentials found in URLs, replacing them with placeholders to prevent accidental exposure.
Audit Metadata