The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides code patterns for fetching and decoding external JSON data from remote APIs (e.g., URLSession.shared.data(from: url) in SKILL.md). This establishes an ingestion point for untrusted data.\n- Ingestion points: Remote API responses fetched via URLSession (SKILL.md).\n- Boundary markers: Absent.\n- Capability inventory: Network operations (URLSession) and structured data parsing (JSONDecoder) across the provided snippets in SKILL.md.\n- Sanitization: Relies on standard Swift type-safety; no explicit sanitization or filtering for downstream prompt usage.

swift-codable