accessibility-dynamic-type-specialist
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions and associated reference documentation are focused on standard Apple accessibility APIs and best practices. It correctly identifies semantic text styles (e.g.,
.body,.headline) and scaling mechanisms (e.g.,UIFontMetrics,@ScaledMetric) as passing patterns, while flagging hardcoded sizes as failures. No evidence of obfuscation, credential theft, or remote code execution was found. - [INDIRECT_PROMPT_INJECTION]: The skill operates on an ingestion surface of untrusted source code provided in the project environment (Step 1 of SKILL.md). While there are no explicit boundary markers or sanitization steps mentioned to handle malicious instructions embedded in comments or code strings, the skill's capability inventory is limited to reading files via
XcodeReadfor auditing purposes. It does not instruct the agent to perform network requests, write to files, or execute shell commands based on the analyzed content, which keeps the risk profile minimal.
Audit Metadata