fix
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various shell commands including
yarn build,yarn lint, andgitoperations. It also heavily utilizes the GitHub CLI (gh) for viewing issues, PRs, and creating new pull requests. - [EXTERNAL_DOWNLOADS]: The instructions explicitly direct the agent to fetch content from external reproduction platforms such as CodeSandbox or StackBlitz when links are found in GitHub issues.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8). It relies on untrusted data from GitHub issues, PR comments, and external site content to guide its implementation and testing logic.
- Ingestion points: Untrusted data enters the agent context via
gh issue view,gh pr view --comments, and external reproduction URLs. - Boundary markers: No specific delimiters or warnings are used to prevent the agent from following malicious instructions embedded within the external data.
- Capability inventory: The skill has broad capabilities including shell command execution, file system modifications, and repository management via Git and GitHub CLI.
- Sanitization: There is no evidence of sanitization or validation of external content before it is used to influence the agent's actions.
Audit Metadata