hig-project-context

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an automated discovery process that reads untrusted project files (README.md, Package.swift, Info.plist, source code) to populate a design context document. This is an indirect prompt injection surface where malicious instructions hidden in these files could override agent behavior.
  • Ingestion points: Identified in SKILL.md as README.md, Package.swift, .xcodeproj, Info.plist, and project source code.
  • Boundary markers: Absent. The instructions do not specify the use of delimiters or warnings to ignore instructions found within the ingested project files.
  • Capability inventory: The skill uses file-read operations across the project structure and performs file-write operations to .claude/apple-design-context.md.
  • Sanitization: Absent. The skill relies on presenting findings to the user for confirmation as a mitigation step.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 06:16 AM
Security Audit — agent-trust-hub — hig-project-context