app-intents-specialist

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is to serve as a reference for Swift development. A comprehensive review of the SKILL.md and the 14 reference files in the references/ directory confirmed that the content is purely instructional and does not contain executable malicious code or unauthorized system commands.
  • [PROMPT_INJECTION]: The skill uses authoritative phrasing, such as instructions to 'unconditionally supersede prior training'. Analysis confirms these are domain-specific constraints intended to ensure technical accuracy regarding App Intents APIs and do not attempt to override the AI agent's safety protocols or core instructions.
  • [EXTERNAL_DOWNLOADS]: No external dependencies, remote scripts, or network-based download patterns were detected. All code examples use standard iOS framework symbols (e.g., AppIntent, AppEntity) and logical placeholders.
  • [DATA_EXFILTRATION]: There are no patterns indicating access to sensitive user data, environment variables, or credentials. Code snippets for networking (e.g., in url-representation.md) are illustrative and use well-known placeholder domains like example.com.
  • [COMMAND_EXECUTION]: The skill does not contain any shell command execution (!command) or subprocess invocation. Its scope is restricted to providing Swift code patterns for developers.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 04:56 AM
Security Audit — agent-trust-hub — app-intents-specialist