do-pr-rev-comment
Pass
Audited by Gen Agent Trust Hub on Jul 19, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: No security issues, malicious code, or suspicious patterns were found in the skill. The content consists only of natural language instructions.
- [PROMPT_INJECTION]: The skill is designed to process PR review comments, which are untrusted external inputs. This constitutes a surface for indirect prompt injection. * Ingestion points: PR review comments. * Boundary markers: No delimiters or instructions are provided to distinguish external data. * Capability inventory: No executable scripts or tool operations are defined in the analyzed file. * Sanitization: No validation or filtering logic is specified for the input data.
- [NO_CODE]: The skill does not contain any scripts, command-line executions, or package dependencies.
Audit Metadata