Skip to content
Security
Command

/evidence-checklist

Generate an evidence preparation checklist for audits or assessments

BOOST
From plugin
claude-grc-plugin
18424 skills1 agent24 commands
Install
> /plugin marketplace add mlunato47/claude-grc-plugin
> /plugin install grc@mlunato47

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/evidence-checklist

Context preview

What this command does when you run it.

Generate an evidence preparation checklist for audits or assessments

Command definition

evidence-checklist.md
description: "Generate an evidence preparation checklist for audits or assessments"

/grc:evidence-checklist

Generate an evidence preparation checklist for upcoming audits or assessments.

Usage

/grc:evidence-checklist [framework] [controls] [audit-type?]

Arguments

  • **framework**: The compliance framework. Accepts: `nist`, `fedramp`, `fisma`, `cmmc`, `soc2`, `iso27001`, `pci`, `hipaa`, `cis`, `cobit`, `ccm`
  • **controls**: Comma-separated control IDs (e.g., `ac-2,ia-2,sc-7`) or a control family (e.g., `ac`, `ia`, `au`)
  • **audit-type** (optional): Type of audit — `3pao`, `annual`, `soc2`, `iso`, `pci`, `internal`. Adjusts evidence focus.

Examples

/grc:evidence-checklist fedramp ac-2,ia-2
/grc:evidence-checklist fedramp ac
/grc:evidence-checklist soc2 CC6.1,CC6.2,CC6.3
/grc:evidence-checklist fedramp ia 3pao
/grc:evidence-checklist pci 8.3,8.4 pci
/grc:evidence-checklist nist cp annual

Behavior

When invoked:

1. **This is a pure reference command** — it generates checklists from framework knowledge and does NOT require user content. No redaction reminder is needed.

2. **Read the OSCAL family JSON first** for authoritative assessment procedures (use the Read tool — do NOT use Bash, Python, or jq to extract data):

  • For `nist`: read `skills/grc-knowledge/oscal/nist-800-53-rev5/{family}.json`
  • For `fedramp`: read `skills/grc-knowledge/oscal/fedramp-moderate-rev5/{family}.json`
  • Extract assessment methods from `.controls[].parts[] | select(.name == "assessment-method")`:
  • **EXAMINE** method → use `.parts[] | select(.name == "assessment-objects") | .prose` as the primary source for "Documents to prepare" and "Technical evidence"
  • **INTERVIEW** method → use `.parts[].prose` as the primary source for "Interview subjects"
  • **TEST** method → use `.parts[].prose` to identify processes and mechanisms that need demonstrable evidence
  • **ID normalization**: `AC-2` → `ac-2`, `AC-2(1)` → `ac-2.1`

3. **Read additional reference files**:

  • Framework file from `skills/grc-knowledge/frameworks/`
  • Audit file from `skills/grc-knowledge/audits/` (if audit-type specified)
  • Narrative quality criteria from `skills/grc-knowledge/audits/narrative-quality-criteria.md` (for evidence expectations)

4. **For each control**, generate:

  • **Documents to prepare**: Policies, plans, procedures, and SSP narratives the auditor will request
  • **Technical evidence**: Screenshots, configuration exports, log samples, scan results
  • **Interview subjects**: Roles the auditor will want to speak with
  • **Common gaps**: Evidence that is frequently missing or insufficient for this control
  • **Preparation tips**: Practical advice for gathering this evidence

5. **If audit-type is specified**, tailor the checklist:

  • `3pao`: Focus on FedRAMP SAR evidence requirements, emphasize technical testing artifacts
  • `annual`: Focus on changes since last assessment, ConMon evidence
  • `soc2`: Focus on operating effectiveness evidence over the observation period
  • `iso`: Focus on documented procedures and management review records
  • `pci`: Focus on scoping documentation and technical validation
  • `internal`: Focus on self-assessment evidence and improvement tracking

6. **If a family is provided** instead of specific controls, expand to all controls in that family at the applicable baseline (default: Moderate for NIST/FedRAMP).

7. **If no arguments provided**, ask the user which framework and controls to generate a checklist for.

Output Format

## Evidence Preparation Checklist: [Framework] [Controls]

[Audit type context if specified]

---

### [Control-ID]: [Title]

**Documents**
- [ ] [Document name — e.g., Access Control Policy]
- [ ] [Document name — e.g., SSP Section 11, AC-2 narrative]

**Technical Evidence**
- [ ] [Evidence item — e.g., Screenshot of account provisioning workflow]
- [ ] [Evidence item — e.g., Export of IAM role assignments]

**Interview Subjects**
- [ ] [Role — e.g., System Administrator (account management process)]
- [ ] [Role — e.g., ISSO (account review process)]

**Common Gaps**
- [Gap description — e.g., Missing evidence of periodic account reviews]

**Tips**
- [Preparation advice]

---

### [Next Control-ID]: [Title]
...

---

## Preparation Timeline

| Timeframe | Action |
|-----------|--------|
| 4 weeks before | [Actions] |
| 2 weeks before | [Actions] |
| 1 week before | [Actions] |
| Day of | [Actions] |

Notes

  • This command generates reference checklists only — it does not require or process user documents.
  • Evidence items are based on what auditors typically request per published assessment procedures (NIST 800-53A, FedRAMP test cases).
  • The checklist format uses markdown checkboxes for easy copy-paste into task trackers.
  • For large families (AC has 25+ controls at Moderate), prioritize the controls most frequently cited in findings.
Read more
Ships withclaude-grc-plugin

A plugin that turns your AI coding agent into a senior GRC (Governance, Risk, and Compliance) analyst. 76+ reference files covering 18 frameworks, 24 slash commands, and deep domain knowledge for federal and commercial compliance work.

Get the whole plugin

Other commands on claude-grc-plugin.