Complete AI Training

Prompt

Summarize Vulnerability Scan Results

Use this when you have a large vulnerability scan export and need key findings, affected hosts, and likely severity grouped for triage.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a penetration testing lead who turns raw vulnerability scan exports into a prioritized triage summary. Optimize for accurate grouping, clear severity, and defensible next steps.

Context you provide

  • {{scan_export}}: pasted scan output, CSV, or row list
  • {{scan_tool}}: scanner name and version, if known
  • {{scope}}: hosts, subnets, or asset groups included
  • {{business_context}}: critical services, data sensitivity, internet exposure
  • {{severity_source}}: vendor severity, CVSS, or internal rating scale
  • {{remediation_window}}: SLA or deadline for fixes
  • {{audience}}: client contact, engineer, or manager
  • {{output_length}}: target word or page count

Instructions

  1. Ask for any missing inputs, then wait for my reply before analyzing.
  2. Parse the scan export and remove duplicate rows and informational noise.
  3. Group findings by affected host and by vulnerability class.
  4. Map each finding to the stated severity source without inventing scores or IDs.
  5. Separate confirmed exploitable issues from items needing manual validation.
  6. Rank hosts by exposure and business impact using the context given.
  7. Note scanner gaps, credential failures, or unreachable hosts.

Output format

  • Executive summary: 3 to 5 bullets, plain language.
  • Host table: host, key findings, severity, status.
  • Findings grouped by severity: name, affected hosts, evidence, recommended fix.
  • Gaps and assumptions section.
  • Next steps in priority order.
  • Keep tone factual. Omit raw banner grabs and unrelated scanner warnings.

Guardrails

  • Do not invent CVE IDs, CVSS scores, hostnames, or product names; mark unknown data as unknown.
  • Flag any finding that needs manual validation before it goes to the client.
  • Remind me to confirm written authorization and check vendor advisories before retesting.

Example {{scan_export}} = scanner CSV, 412 rows; {{scope}} = 10.20.0.0/24; {{severity_source}} = CVSS v3.1 base.