qa
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
ghcommand-line tool to perform issue creation on GitHub (gh issue create). - [DATA_EXFILTRATION]: Data derived from user descriptions and codebase exploration is transmitted to GitHub. The skill is designed to file these issues immediately without an intermediate user confirmation step, but it targets a well-known service as part of its primary functionality.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes external codebase data to generate issue bodies. \n
- Ingestion points: Reads project files such as
UBIQUITOUS_LANGUAGE.mdand performs automated codebase exploration (SKILL.md). \n - Boundary markers: No explicit delimiters or instructions are used to separate codebase data from the agent's core logic. \n
- Capability inventory: Uses the
ghtool to write data to an external repository (SKILL.md). \n - Sanitization: The skill includes instructions to omit file paths, line numbers, and internal implementation details, serving as a functional constraint on the generated output.
Audit Metadata