skills/notque/vexjoy-agent/swift/Gen Agent Trust Hub

swift

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process untrusted user-provided code and project structures during its execution phases.
  • Ingestion points: User code and project files are processed in Phase 2 (EXECUTE) and verified in Phase 3 (VERIFY).
  • Boundary markers: No explicit markers or instructions are provided to the agent to treat user-provided code as untrusted data or to ignore embedded instructions.
  • Capability inventory: The skill uses the shell to execute swift build and swift test --enable-code-coverage (SKILL.md).
  • Sanitization: There is no evidence of input validation or sanitization for user-provided code or build configurations like Package.swift.
  • [COMMAND_EXECUTION]: The skill explicitly instructs the agent to run build and test commands. While these are standard developer tools, running them on untrusted projects poses a risk because Swift's build system (SwiftPM) and test runners can execute arbitrary code defined in the project's manifest or test cases.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 08:53 PM
Security Audit — agent-trust-hub — swift