Skills
Navigate
skills/rmyndharis/antigravity-skills/multi-platform-apps-multi-platform/Gen Agent Trust Hub

multi-platform-apps-multi-platform

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill interpolates untrusted user input into prompts for specialized sub-agents and chains their outputs without adequate sanitization or boundary markers.
  • Ingestion points: The $ARGUMENTS variable in SKILL.md captures initial feature specifications which are directly interpolated into the first task prompt for the backend-architect.
  • Boundary markers: Instructions in SKILL.md do not utilize delimiters (such as triple backticks) or specific safety warnings to isolate user input from system instructions.
  • Capability inventory: The skill uses the Task tool to invoke specialized agents (including backend-architect, ios-developer, and unit-testing::test-automator) across SKILL.md to generate code and documentation.
  • Sanitization: There is no evidence of validation or filtering for input passed through $ARGUMENTS in the provided SKILL.md file.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 12:36 AM