Skip to main content
VISO Chat Agent is designed to understand natural language, but the quality of its answers depends heavily on how clearly you phrase your request. These tips will help you get more accurate, specific, and useful responses.

​
Be specific about what you want

Vague questions produce vague answers. The more precisely you describe what you’re looking for, the more targeted the response.

​
Name the vendor or entity

For vendor-specific questions, always include the vendor’s name. VISO Chat Agent can narrow its search significantly when it knows exactly which relationship you’re asking about.
“What is Salesforce’s current residual risk score?” “Does Workday have a current ISO 27001 certificate on file?” “Who is the business owner for the Stripe relationship?”

​
Break complex questions into steps

If you’re working through a multi-step decision, break it into a sequence rather than asking everything at once. VISO Chat Agent handles focused questions better than compound ones. Instead of: “Show me all vendors with High risk who haven’t responded to my recent advisory outreach and need a follow-up questionnaire and are in the Engineering BU” Try:
  1. “Which vendors in Engineering have High or Extreme residual risk?”
  2. “Of those, which have an active advisory outreach request with no response?”
  3. “For those vendors, what follow-up questionnaires have been sent?”

​
Use the contextual prompts

When VISO Chat Agent surfaces a suggested prompt based on what you’re looking at (like an advisory page or an assessment result), use it. These prompts are pre-tuned to the context and often lead to better results than starting from scratch.

​
Specify filters when you have them

If you’re interested in a specific slice of your portfolio, say so. VISO Chat Agent respects filters like:
  • Business unit (“in the Legal business unit”)
  • Vendor tier (“Tier 1 vendors”)
  • Tags (“vendors tagged ‘critical infrastructure’”)
  • Risk level (“vendors with Extreme inherent risk”)
  • People, by name or email (“where Jane Doe is the assessment lead”)
  • Recertification (“with an overdue recertification”)
  • Current audit reports (“with a current SOC 2 Type II”)
  • Nth parties (“vendors that use AWS”)
  • Time range (“in the last 90 days”)
Use the names exactly as they appear in VISO TRUST. If a name doesn’t match a user, tier, or nth party in your organization, VISO Chat Agent returns no matches for that filter. It doesn’t drop the filter and widen the search to all relationships.

​
Ask for the underlying evidence

When VISO Chat Agent gives you an answer about a vendor’s security posture, you can ask it to show the supporting detections:
“What’s the source for that?” “Which artifact does that come from?” “Show me the detection reference for the incident response control.”
This is especially useful when reviewing answers in a vendor relationship. You can verify that the answer is grounded in real, analyzed documentation rather than a general assumption.

​
If the answer seems wrong

  • Rephrase the question — try describing what you’re looking for from a different angle
  • Add more context — include the vendor name, business unit, time range, or specific status
  • Check the source references — VISO Chat Agent cites its sources. If the citation doesn’t match your expectation, the underlying data may need updating (e.g., a relationship context that hasn’t been configured, or an artifact that hasn’t been uploaded yet)
  • Break it into smaller questions — start with a simpler version of your question and build from there
AI-generated answers can be incorrect or incomplete even when the underlying data is correct. Compare the answer with the relationship and supporting evidence, check dates and scope, and contact support if they disagree. See How VISO TRUST Uses AI for guidance.