The Agent Skills Directory

[SAFE]: The skill is focused purely on software engineering tasks, specifically the modernization of UIKit and SwiftUI codebases. It provides clear, non-malicious logic for API substitution and UI configuration.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided project files and source code as input. This creates a surface for indirect prompt injection if a processed file contains malicious instructions intended to mislead the agent performing the modernization.
Ingestion points: Local source code files (.swift, .m, .h) and configuration files (Info.plist, build settings) identified during the detection phase (SKILL.md).
Boundary markers: The instructions do not specify the use of clear delimiters when reading or writing code, which is typical for this type of refactoring tool.
Capability inventory: The skill possesses file-write and project-configuration-modification capabilities (e.g., creating SceneDelegate.swift, updating Info.plist).
Sanitization: No explicit sanitization or filtering of input code content is described beyond standard pattern matching for deprecated APIs.

app-resizability