deep-module-refactor
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates the creation of GitHub issues using the
gh issue createcommand. It explicitly instructs the agent to bypass user review before submission ("Do NOT ask the user to review before creating"), which reduces user control over external actions. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from a codebase to generate architectural designs and GitHub issues.
- Ingestion points: Reads codebase content using the
Agenttool withsubagent_type=Exploreas described inSKILL.md. - Boundary markers: None identified. The skill does not define clear delimiters or instructions to ignore embedded commands within the processed source code.
- Capability inventory: The skill has the ability to spawn multiple sub-agents and execute
gh issue createto post content to external repositories. - Sanitization: There is no evidence of sanitization or validation of the code content before it is used to frame problem spaces or populate GitHub issue templates.
Audit Metadata