user-research-synthesis

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of markdown documentation providing guidelines and methodologies for qualitative and quantitative research synthesis. It does not perform any technical operations or access external resources.
  • [NO_CODE]: No scripts, binaries, or software packages are included or referenced in the skill definition.
  • [PROMPT_INJECTION]: The skill is designed to process untrusted external data, such as interview notes and survey responses, creating a surface for indirect prompt injection.
  • Ingestion points: Raw qualitative data, interview notes, survey responses, support tickets, and behavioral data (SKILL.md).
  • Boundary markers: No specific boundary markers or 'ignore' instructions are used to separate user data from system instructions.
  • Capability inventory: The skill has no functional capabilities (e.g., file system access, network requests, or code execution) across its documentation.
  • Sanitization: No input validation or sanitization mechanisms are defined to handle malicious instructions embedded in research data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 08:21 PM