Skills
Navigate
skills/nexu-io/open-design/research-decision-room/Gen Agent Trust Hub

research-decision-room

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user research data to generate a decision-making artifact.
  • Ingestion points: The research_material input in SKILL.md is the primary entry point for external data.
  • Capability inventory: The skill utilizes file_write to output the final index.html file.
  • Boundary markers: The workflow includes normalization and validation steps, such as using 'not provided' for missing data, to structure the input signal.
  • Sanitization: The evidence-model.md provides integrity rules for handling quotes and evidence strength to minimize misinterpretation of external data.
  • [DYNAMIC_EXECUTION]: The skill generates a report containing client-side JavaScript for interactivity.
  • Evidence: Step 6 of SKILL.md instructs the agent to include vanilla JavaScript for filtering evidence and switching views within the generated HTML artifact.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:59 PM
Security Audit — agent-trust-hub — research-decision-room