motor-control
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [NO_CODE] (SAFE): The analyzed file is a documentation-only skill definition with no associated scripts, binaries, or configuration files that execute commands.
- [Prompt Injection] (SAFE): Instructions focus on grounding the agent's output in specific reference files and do not attempt to bypass core safety guardrails or extract system prompts.
- [Indirect Prompt Injection] (LOW): The skill ingests content from reference files (e.g., patterns.md) which could theoretically contain instructions. However, the skill lacks external network or file-write capabilities, meaning any injection would be confined to the conversation context (display only).
- [Data Exposure] (SAFE): No patterns for accessing sensitive files or hardcoded credentials were found.
Audit Metadata