swiftlint
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill actively promotes secure development practices by advising users to review auto-fixes before committing and providing explicit warnings about the risks of using CI flags like -skipPackagePluginValidation that bypass security checks for external plugins.- [EXTERNAL_DOWNLOADS]: The skill includes instructions for integrating standard third-party tools, such as the SwiftLintPlugins package and the Homebrew distribution of SwiftLint. It mitigates supply chain risks by recommending version pinning and reviewing these external dependencies before use.- [INDIRECT_PROMPT_INJECTION]: The skill documents the processing of external Swift projects and configuration files, which defines an indirect prompt injection surface.
- Ingestion points: Project directories containing .swift files and .swiftlint.yml configuration files.
- Boundary markers: File analysis is constrained by user-defined included and excluded path configurations.
- Capability inventory: The agent can invoke the swiftlint CLI, which may modify files when using --fix or perform network requests if parent_config is set to a URL.
- Sanitization: The skill advocates for manual review of all tool-generated changes and incremental adoption via baselines to ensure the agent's output is verified by a human.
Audit Metadata