The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill documents patterns for ingesting and processing untrusted data, which constitutes a potential attack surface.
Ingestion points: Data enters the context through searchText in SearchViewModel, service.getUsers() in HomeViewModel, and URLSession.shared.dataTaskPublisher in SKILL.md.
Boundary markers: The provided code templates do not include explicit delimiters or sanitization logic to distinguish between data and instructions.
Capability inventory: The skill demonstrates usage of URLSession for network operations and DispatchQueue for task scheduling.
Sanitization: There is no evidence of input validation or output escaping in the provided Swift code examples.
[EXTERNAL_DOWNLOADS]: The documentation references an external testing library.
The skill recommends using pointfreeco/combine-schedulers for testing time-based operators, which is a recognized resource in the Swift development community.

combine-framework