research-ops
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed entirely of natural language instructions guiding the agent on how to categorize and process information. There are no executable scripts, shell commands, or external code dependencies included.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and normalize user-provided evidence and external research data. It includes defensive instructions for the agent to maintain explicit 'evidence boundaries' and to separate sourced facts from user-supplied context and inferences, which mitigates the risk of instructions embedded in data being misread as system directives.
Audit Metadata