qa
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted user descriptions of bugs and uses them to populate the body of GitHub issues.
- Ingestion points: User-provided problem descriptions and codebase content retrieved by the background exploration agent (SKILL.md).
- Boundary markers: The skill provides specific templates for issue creation but does not explicitly instruct the agent to ignore or sanitize potentially malicious instructions embedded within the user's bug report.
- Capability inventory: The agent utilizes the
gh issue createtool to write data to external repositories and uses a subagent to read local codebase files. - Sanitization: The skill includes instructions to omit file paths and implementation details, which helps maintain privacy, but it lacks specific mechanisms to prevent the user from injecting commands into the issue creation process.
Audit Metadata