soia-dev-design-explorer
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts, such as
detect_route.py,check_env.py, anddaemon_ctl.py, to manage the execution environment and communicate with Open Design daemons. These scripts are located within the directory of its mandatory dependency,soia-dev-open-design-ops(~/.agents/skills/soia-dev-open-design-ops/scripts/). - [EXTERNAL_DOWNLOADS]: The documentation provides installation commands for the skill and its dependencies using
npxandclaude pluginfrom the author's officialsoia-teamrepositories on GitHub. - [PROMPT_INJECTION]: The skill implements a workflow that inlines the content of external design files, such as
tokens.cssandDESIGN.md, directly into the prompts used by the Open Design generation agent. While this is a functional requirement for maintaining design consistency, it represents a surface for indirect prompt injection if the ingested design assets originate from untrusted sources. - [SAFE]: All identified operations, including script execution and dependency management, are consistent with the skill's stated purpose and involve resources managed by the verified author (soia-team).
Audit Metadata