issues
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data for GitHub issue titles, descriptions, and labels.
- Ingestion points: Interaction steps in SKILL.md collect user-defined titles, bodies, and labels.
- Boundary markers: Use of HEREDOC is suggested for issue bodies, but other command-line arguments like
--titleand--labellack delimiters. - Capability inventory: The skill executes
ghcommands via theBashtool, which can run arbitrary shell commands if strings are not properly quoted or escaped. - Sanitization: No explicit sanitization or escaping for shell metacharacters is instructed for the title or label inputs.
- [COMMAND_EXECUTION]: The skill executes
gh issue create,gh issue list, andgh issue viewcommands. These commands are run in a bash environment and take parameters derived directly from user input.
Audit Metadata