ptlam-code-style-swift

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute standard Swift development commands for code verification and maintenance.
  • Commands include swift build, swift test, swiftformat, and swiftlint as part of the normal development workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by processing untrusted project files while possessing execution capabilities.
  • Ingestion points: The agent is instructed to read AGENTS.md, Package.swift, .swiftformat, .swiftlint.yml, Package.resolved, and all source files in scope.
  • Boundary markers: No explicit instructions to ignore or delimit instructions found within those data files are provided.
  • Capability inventory: The agent utilizes shell execution capabilities via the swift toolchain (build, test, format, lint).
  • Sanitization: There is no explicit sanitization or filtering logic described for the content ingested from the project files.
  • [DATA_EXPOSURE]: The skill promotes secure logging practices in the foundation reference.
  • The logging.md reference explicitly mandates that developers "Never log a secret or a person," providing specific examples of data to exclude such as credentials, tokens, and PII.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 06:25 PM
Security Audit — agent-trust-hub — ptlam-code-style-swift