design-tokens
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python scripts:
python3 .claude/skills/_ux-ui-shared/scripts/contrast.pyandpython3 .claude/skills/_ux-ui-shared/scripts/validate_tokens.py. These are internal project scripts used for design token validation and accessibility auditing. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local configuration files and project documentation which could potentially contain embedded instructions.
- Ingestion points:
CLAUDE.mdand various JSON files in the.claude/skills/_ux-ui-shared/tokens/directory. - Boundary markers: None explicitly defined to separate external data from agent instructions.
- Capability inventory: Ability to execute local Python commands and read/write project files.
- Sanitization: None specified for the ingested file contents.
Audit Metadata