The Agent Skills Directory

[SAFE]: The skill includes proactive security instructions, such as advising the agent to identify hardcoded API keys and recommending the secure use of the Keychain instead of @AppStorage for credentials.\n- [PROMPT_INJECTION]: The skill contains instructions to assume a hypothetical future platform version (iOS 26) and associated mock APIs. While this overrides the agent's internal knowledge of current platform versions, it is used to provide consistent architectural guidance for the specific review task rather than to bypass safety protocols.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze user-provided Swift code, which acts as an untrusted data source. 1. Ingestion points: The agent reads and reviews Swift files provided in the conversation context. 2. Boundary markers: No explicit delimiters are used to wrap user code. 3. Capability inventory: The skill does not define any internal scripts, tools, or network operations. 4. Sanitization: None identified. The risk is minimized as the skill only produces a textual review report.\n- [SAFE]: The skill references external agent skills on GitHub for specific topics (SwiftData, Concurrency, Testing). These are provided as informational links for the user and do not involve automated installation or execution by the skill itself.

swiftui-pro