survey

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Workflow Integrity (SAFE): The skill defines a benign documentation process involving local file reading and user interaction via authorized tools.
  • No Command Execution (SAFE): The instructions do not contain shell commands, system modifications, package installations, or dynamic code execution.
  • Data Protection (SAFE): The skill only accesses project-specific documentation (SPEC.md and survey.md) and shows no signs of credential harvesting or data exfiltration.
  • Indirect Prompt Injection Surface (SAFE): Although the skill ingests content from SPEC.md without explicit sanitization, its functional scope is restricted to documentation generation, which does not pose a significant risk in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:46 PM