skills/srinitude/skills/dtcg-tokens/Gen Agent Trust Hub

dtcg-tokens

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches typography metadata and font assets from Google Fonts official domains (fonts.google.com, fonts.gstatic.com) and official GitHub repositories to embed them into offline artifacts. This is a documented part of its design-token generation workflow and targets trusted services.
  • [COMMAND_EXECUTION]: Python's subprocess module is utilized within the test suite (scripts/tests/) to verify the functionality of internal scripts. These calls are restricted to the skill's own scripts and use safe argument list patterns for execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted multimodal data to drive its design decisions, which is a potential injection surface. The risk is mitigated by a rigorous 25-step state machine and a dedicated contract-checking script (scripts/lib/artifact_contract.py) that audits the authored HTML candidate for unauthorized assets, scripts, or forbidden markers before final assembly.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 02:24 PM
Security Audit — agent-trust-hub — dtcg-tokens