rfc-status
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. Ingestion points: RFC files in the 'rfcs/' directory and user inputs via '--comment' or '--reviewer'. Boundary markers: Absent. Capability inventory: File-write (updating frontmatter) and command-execution (git mv). Sanitization: Absent. Malicious instructions within an RFC file or user-provided comment could potentially influence agent behavior during processing.
- [COMMAND_EXECUTION]: The skill utilizes the 'git mv' command to move files between directories based on status changes. While appropriate for the task, executing commands with parameters derived from external files or user input without explicit validation poses a potential risk if not handled securely by the underlying agent implementation.
Audit Metadata