The Agent Skills Directory

[SAFE]: The skill provides high-quality documentation and code implementation patterns for iOS networking that align with security best practices. All external links point to documentation resources.
[DATA_EXFILTRATION]: The LoggingMiddleware implementation in references/resilience-security-and-streaming.md includes logic to explicitly redact the Authorization header, preventing sensitive credentials from being exposed in debug logs.
[CREDENTIALS_UNSAFE]: The evaluation requirements in evals/evals.json correctly identify that sensitive tokens should be stored in the Keychain rather than UserDefaults, promoting secure credential management.
[INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing external API data. 1. Ingestion points: APIClient.request (via URLSession.data(for:)) in references/api-client-and-request-building.md. 2. Boundary markers: The implementation includes HTTP status code validation (200-299) to verify response integrity. 3. Capability inventory: The skill demonstrates network tasks (URLSession) and local file system operations (FileManager). 4. Sanitization: Data is parsed using structured JSONDecoder interfaces, which mitigates raw data injection risks into application logic.

ios-networking