swift-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill contains documentation and instructional patterns for manual code review. It does not include any automated scripts, shell command executions, or network operations. All external references target well-known domains for documentation purposes.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted Swift source code. (1) Ingestion Point: External .swift files provided for review. (2) Boundary Markers: The skill employs a specific output format [FILE:LINE] to demarcate findings, although explicit input delimiters for the source code are not defined. (3) Capabilities: The skill's functionality is limited to textual analysis and report generation. (4) Sanitization: None present. The risk is inherent to the code review use case and is mitigated by the lack of high-privilege capabilities.
Audit Metadata