github-issue-creator
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the standard
ghCLI tool to create issues. This is the intended behavior and relies on the user's local authentication. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data to generate GitHub issues. Ingestion points: User-provided issue details (SKILL.md). Boundary markers: None explicitly defined for the issue body. Capability inventory: Executes
gh issue createandrm, and reads local templates (SKILL.md). Sanitization: The skill mitigates shell injection by requiring the body to be written to a temporary file and passed via--body-file.
Audit Metadata