compliance-gap-analyzer
Aggregates audit findings mapped to framework controls, classifies each control, identifies blind spots, prioritizes gaps by severity, and produces a complete gap report supports ISO 27001, NIST CSF, PCI-DSS, and OWASP. Trigger when the user provides audit findings tagged to framework controls and asks for a gap report, asks what controls are failing or what their biggest compliance gaps are, needs to identify blind spots or untested controls, wants to compare current compliance posture against a previous audit, or shares pentest/audit results wanting them mapped to a compliance framework.
Rifteo · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add Rifteo/skills --skill compliance-gap-analyzer技能说明
name: compliance-gap-analyzer description: Aggregates audit findings mapped to framework controls, classifies each control, identifies blind spots, prioritizes gaps by severity, and produces a complete gap report supports ISO 27001, NIST CSF, PCI-DSS, and OWASP. Trigger when the user provides audit findings tagged to framework controls and asks for a gap report, asks what controls are failing or what their biggest compliance gaps are, needs to identify blind spots or untested controls, wants to compare current compliance posture against a previous audit, or shares pentest/audit results wanting them mapped to a compliance framework. license: MIT metadata: version: "0.0.1" author: Rifteo tags: ["compliance", "audit", "gap-analysis", "iso27001", "nist-csf", "pci-dss", "owasp"]
Compliance Gap Analyzer
When an auditor provides findings from an audit — each tagged to a framework control with a compliance status — aggregate them into a structured gap report. Identify what is failing, what was never tested, and what to fix first.
When NOT to use
Do NOT activate this skill — respond normally — when:
- The user is asking about a single vulnerability or finding (use
finding-writerorcvss-scorerinstead) - The user wants exploit information or attack methodology (use
check-exploitinstead) - The user is asking what a specific control means or requires (use
control-lookupinstead) - The user is asking a general compliance question without providing actual audit findings
- No findings list has been provided — ask for findings before producing any gap report
Step 1 — Gather Inputs
Before analyzing, confirm you have:
- Framework(s) in scope — ISO 27001, NIST CSF, PCI-DSS, OWASP Top 10, OWASP ASVS, or a custom framework
- Findings list — for each finding:
- Control reference (e.g., ISO 27001 A.9.1.1, NIST CSF PR.AC-1, PCI-DSS Req 6.3, OWASP A03:2021, OWASP ASVS V2.1.1)
- Compliance status: Compliant, Partially Compliant, Non-Compliant, or Not Tested
- Severity of the finding: Critical, High, Medium, Low, or Informational
- Audit scope — which parts of the organization, systems, or processes were in scope
- Previous audit gap state (optional) — a prior gap report to enable delta analysis
If findings are provided as raw notes or a table, normalize them before proceeding. If the framework is not specified, infer it from the control references used.
Multi-framework findings: a single finding may map to controls in multiple frameworks simultaneously. Include it in the gap analysis for each framework it references — do not drop cross-mapped findings.
Step 2 — Normalize and Group
- Group findings by framework, then by control reference within each framework.
- Normalize control IDs to their canonical notation (e.g., "Access Control 9.1" → "ISO 27001 A.9.1").
- For each control, collect all findings assigned to it and their compliance statuses.
Step 3 — Classify Each Control
Apply the weakest-link rule: one non-compliant finding is enough to classify the entire control as non-compliant.
| Classification | Criteria |
|---|---|
| Compliant | All findings for this control pass; no gaps identified |
| Partially Compliant | Mix of compliant and partially compliant findings; control is not fully satisfied |
| Non-Compliant | One or more findings explicitly fail this control |
| Not Tested | No findings assigned to this control — it was not audited (blind spot) |
A control rated Partially Compliant must be treated as a gap — partial compliance is not compliance.
Step 4 — Prioritize Gaps
Assign a gap priority to each non-compliant or partially compliant control based on two factors:
Finding severity (worst finding assigned to that control):
| Severity | Weight |
|---|---|
| Critical | 4 |
| High | 3 |
| Medium | 2 |
| Low | 1 |
| Informational | 0 |
Control criticality within the framework:
| Criticality | Weight | Examples |
|---|---|---|
| Foundational | +2 | Access control, identity management, encryption at rest/transit, incident response |
| Standard | +1 | Most operational and procedural controls |
| Supporting | 0 | Documentation, awareness, supplemental controls |
Gap Priority Score = Finding Severity Weight + Control Criticality Weight
| Score | Gap Priority |
|---|---|
| 5–6 | Critical — remediate within 48 hours |
| 3–4 | High — remediate within 7 days |
| 2 | Medium — remediate within 30 days |
| 0–1 | Low — schedule in next maintenance cycle |
Blind spots (Not Tested controls) are always flagged separately — they carry unknown risk and must be acknowledged regardless of priority score.
Step 5 — Delta Analysis (only if previous audit state is provided)
Compare the current gap state against the previous audit:
| Change Type | Definition |
|---|---|
| New gap | Control was Compliant or Not Tested before; now Non-Compliant or Partially Compliant |
| Closed gap | Control was Non-Compliant or Partially Compliant before; now Compliant |
| Persistent gap | Control was Non-Compliant or Partially Compliant before; still is |
| Regression | Control was Compliant before; now Non-Compliant (flag explicitly — this is a priority signal) |
| New blind spot | Control was tested before; now Not Tested — scope was reduced |
| Blind spot resolved | Control was Not Tested before; now has findings |
Regressions must always be called out prominently — a control that passed before and now fails indicates a process breakdown, not just a gap.
Output Format
Produce exactly this structure. Do not omit any section. If a section has no entries, state "None."
Audit: [Audit name or identifier, if provided] Framework(s): [List all frameworks analyzed] Scope: [Systems, processes, or organizational units in scope] Findings analyzed: [Total count] Controls assessed: [Count of contro
阅读完整来源 ↗