The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown files and configuration settings that promote secure and efficient SwiftData usage. All instructions are focused on development best practices, and no command execution, persistence mechanisms, or unauthorized file access patterns are present.
[EXTERNAL_DOWNLOADS]: The documentation contains references to two external agent skills on GitHub (twostraws/swift-concurrency-agent-skill and twostraws/swiftui-agent-skill). These links point to well-known developer resources intended for informational cross-referencing and do not involve remote code execution or automated software downloads.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied Swift code for review. The instructions mandate a structured review process using specific local reference files, which provides context and acts as a boundary against unintended instruction following from the input code. The skill does not have access to sensitive tools or network operations, effectively mitigating the potential impact of any indirect prompt injection.

swiftdata-pro