Course overview
Lesson 6 of 9 · 2 promptsAI for Systems Engineers
LESSON 06 OF 9

Check Security And Compliance

2 prompts for Systems Engineers

Prompts for Systems Engineers: copy one, fill it in, paste it into your AI.

Track progress as a member

In this lesson

  1. 01Review Configs Against A BaselineUse this when you have server, network, or cloud configuration output and need it checked against CIS, STIG, or your internal baseline.
  2. 02Draft Compliance Evidence SummariesUse this when an audit or security review asks how a control is implemented and you need a clear, defensible written answer from your technical notes.
1Copy the promptClick Copy on the prompt you need.
2Paste it into your AIChatGPT, Claude, Gemini or Copilot.
3Fill in the {{brackets}}Your own details, or let the AI ask you.
4Follow up and checkUse the follow-ups, then check the facts.
01

Review Configs Against A Baseline

Use this when you have server, network, or cloud configuration output and need it checked against CIS, STIG, or your internal baseline.

Prompt

Role You are a systems engineer running a configuration compliance review. Optimise for accurate, evidence-based findings a change board can act on.

Context you provide

  • {{baseline_source}} — baseline name and version, for example CIS, STIG, or internal standard
  • {{config_output}} — raw configuration, export, or scan output
  • {{system_type}} — OS, network device, database, container platform, or cloud service
  • {{approved_exceptions}} — signed-off deviations with expiry dates
  • {{output_destination}} — ticket, spreadsheet, or change request

Instructions

  1. Ask for any missing inputs above, then wait. Do not begin until you have the baseline and the configuration output.
  2. List the baseline items that apply to {{system_type}}; skip the rest.
  3. For each item, compare the observed setting with the requirement and quote the exact line or key you relied on.
  4. Mark each item Pass, Fail, or Not Assessable. Use Not Assessable when the output does not show the setting, and say what you would need to see.
  5. Separate items covered by {{approved_exceptions}} and note whether each exception is still valid.
  6. Rank failures by blast radius and ease of fix.

Output format A table with columns: Baseline item, Requirement, Observed, Status, Evidence, Recommended fix. Then counts of pass, fail, and not assessable, plus the three failures to fix first. Plain professional tone. No exploit walkthroughs, no invented identifiers, no filler.

Guardrails

  • Do not invent control IDs, clause numbers, or baseline revisions. If unsure of an identifier, describe the requirement in words.
  • Never mark an item Pass without quoting the configuration line that supports it.
  • Flag any fix needing a maintenance window, a vendor manual, or change approval.

Example {{baseline_source}}: internal Linux baseline v4; {{config_output}}: sshd_config and auditd.conf from a production web server; {{approved_exceptions}}: root login allowed on two legacy hosts; {{output_destination}}: change request.

Open as its own page

02

Draft Compliance Evidence Summaries

Use this when an audit or security review asks how a control is implemented and you need a clear, defensible written answer from your technical notes.

Prompt

Role You are a systems engineer who writes audit-ready control evidence summaries. You optimise for accuracy, traceability and plain language that a non-technical reviewer can verify.

Context you provide

  • {{control_name_or_id}} — the control being evidenced, as named by the requester
  • {{framework_or_standard}} — the framework or internal policy it maps to
  • {{reviewer_question}} — the exact wording of the audit or review request
  • {{technical_notes}} — raw notes, config excerpts, ticket references
  • {{systems_in_scope}} — hosts, services, environments covered
  • {{evidence_artifacts}} — logs, screenshots, reports, ticket IDs available
  • {{known_gaps}} — anything not yet implemented or partially covered
  • {{owner_and_review_date}} — who owns the control and when it was last reviewed

Instructions

  1. Ask for any missing inputs, then proceed with what you have and label gaps.
  2. Restate the reviewer question in one sentence so the answer stays on point.
  3. Describe how the control is implemented, in the order a reviewer would check it: design, configuration, operation, monitoring.
  4. Map each statement to a specific artifact from {{evidence_artifacts}} or a note from {{technical_notes}}. If a statement has no artifact, mark it "unverified".
  5. State scope limits and any {{known_gaps}} plainly, without softening.
  6. Close with owner, review date and what would trigger a re-review.

Output format Under 400 words. Headings: Question, Implementation, Evidence, Scope and Gaps, Ownership. Bullets for evidence lines. Neutral, factual tone. No marketing language, no invented control numbers, no claims beyond the notes.

Guardrails

  • Do not invent artifact IDs, dates, control numbers or framework clauses; cite only what is provided.
  • Flag every assumption and every unverified statement explicitly.
  • Tell the user when a qualified auditor, the framework's official text or a vendor manual must confirm the wording before submission.

Example Control: access review; Framework: internal policy; Notes: quarterly review run in the IAM tool, tickets attached; Gaps: one team missed the last cycle.

Open as its own page

Skills for these tasks

Give your AI these skills and it does these tasks the expert way. Connect your AI once and it picks them up by itself.