The Agent Skills Directory

[SAFE]: The skill performs legitimate repository analysis to determine task types (symbols, articles, etc.) and repo shapes (Swift Package or Xcode). It provides clear handoff instructions for tasks outside its primary scope, such as building or searching documentation.
[COMMAND_EXECUTION]: The skill includes Python scripts (scripts/run_workflow.py and scripts/customization_config.py) that are executed via the uv tool. Analysis of these scripts shows they perform safe operations like file existence checks, directory globbing, and safe YAML parsing. No arbitrary command execution or shell injection vectors were identified.
[EXTERNAL_DOWNLOADS]: The skill references documentation and source code from official and well-known sources, including apple.com, swift.org, and github.com/swiftlang. These references are documented for user guidance and do not represent a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill ingests a request input to infer task types and repo shapes.
Ingestion points: request parameter in scripts/run_workflow.py.
Boundary markers: Absent.
Capability inventory: File system reads in scripts/run_workflow.py and file system writes for configuration in scripts/customization_config.py.
Sanitization: Input is normalized (lowercase, whitespace stripping) for string matching. The risk is negligible as the input does not flow into execution or system-altering sinks.

author-swift-docc-docs