The Agent Skills Directory

[SAFE]: The skill provides patterns and instructions for migrating Swift code to modern iOS 17/18 features. It operates as a guidance tool for the agent and does not execute any code.
[DATA_EXFILTRATION]: The skill instructs the agent to search project Swift files but does not include any network operations or access to sensitive credentials files.
[PROMPT_INJECTION]: 1. Ingestion points: Local Swift source files (**.swift). 2. Boundary markers: Absent. 3. Capability inventory: No file write, shell execution, or network capabilities are present; model invocation is disabled. 4. Sanitization: Absent. The skill is safe against indirect prompt injection because the agent is restricted from performing any actions that could be exploited by malicious content in the source files.

axiom-modernize