@6171/experiment-research-lab

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for searching and documenting scientific experiments, utilizing standard tools for research and content generation.
  • [PROMPT_INJECTION]: The skill processes untrusted information from the web to formulate experiment plans, presenting a surface for indirect prompt injection.
  • Ingestion points: Search results from GitHub, Google Scholar, and maker forums (SKILL.md Phase 2).
  • Boundary markers: None specified for handling external search content.
  • Capability inventory: Generates firmware code (.ino), Python scripts, and documentation files.
  • Sanitization: There are no specific instructions for sanitizing or validating search results before implementation.
  • [COMMAND_EXECUTION]: The skill utilizes the tar command to bundle project artifacts, which is an expected and benign operation within its described workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 05:15 AM
Security Audit — agent-trust-hub — @6171/experiment-research-lab