PR

prowler-compliance-review

Validates compliance framework JSON files and reviews PRs for security standards.

Install

mkdir -p .claude/skills/prowler-compliance-review && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/7923" && unzip -o skill.zip -d .claude/skills/prowler-compliance-review && rm skill.zip

Installs to .claude/skills/prowler-compliance-review

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Reviews Pull Requests that add or modify compliance frameworks. Trigger: When reviewing PRs with compliance framework changes, CIS/NIST/PCI-DSS additions, or compliance JSON files.
180 chars · catalog description✓ has a “when” trigger
Advanced

Key capabilities

  • →Validate JSON syntax for compliance framework files
  • →Check for missing or duplicate security requirement IDs
  • →Verify framework metadata completeness
  • →Compare new dashboard files against established patterns
  • →Test compliance frameworks using Prowler CLI

How it works

The skill uses a validation script to check JSON structure and requirement IDs, followed by a manual review process for changelogs and dashboard patterns.

Inputs & outputs

You give it
Compliance framework JSON file
You get back
Validation status and approval decision

When to use prowler-compliance-review

  • →Validate JSON schema for compliance frameworks
  • →Check for missing or duplicate security requirement IDs
  • →Compare new framework dashboards with existing patterns
  • →Automate PR reviews for CIS or NIST updates

About this skill

When to Use

  • Reviewing PRs that add, extend or modify compliance frameworks under prowler/compliance/
  • Reviewing check-mapping changes, ConfigRequirements guardrails, or new compliance formatters

For authoring rules and the schema, load prowler-compliance.

Identify the schema first

FileSchemaExpect in the PR
prowler/compliance/<key>.json (top level, lowercase keys)UniversalJSON only (+ optional UI mapper/panel/icon). No Python formatter, no COMPLIANCE_CLASS_MAP entry
prowler/compliance/<provider>/<key>.json (Capitalized keys)LegacyJSON; plus formatter + __main__.py branch + COMPLIANCE_CLASS_MAP entry only if a new framework family/provider writer is introduced

A brand-new framework submitted as legacy is a REQUEST CHANGES unless it extends an existing legacy family.

Review Checklist

CheckHowPass criteria
Loadsload_compliance_framework_universal() one-liner belowNot None (universal loader swallows errors, check the log)
Catalog integrityuv run pytest tests/lib/check/compliance_catalog_integrity_test.py -k <basename>Unique ids, no repeated checks, every check exists for its provider
Structurevalidate_compliance.py <file>No errors (both schemas)
outputs.table_config (universal)validatorPresent, group_by declared in attributes_metadata
attributes_metadata (universal)readPresent, required/enum used for closed sets
GuardrailsreadUniversal: Provider on every config_requirements entry. Value = strictest the control allows; ConfigKey exists in the check
VersionreadSet for versioned frameworks; filename version equals Version
Mapping qualitysample requirements vs check titlesCheck literally verifies the requirement; otherwise [] (manual). No padding
Completenesscompare with sourceEvery source requirement present, including manual ones
Grouping valuesjq '[.Requirements[].Attributes[0].Section] | unique' (legacy)No spelling variants (they split the UI tree)
Testsuv run pytest -n auto tests/lib/check/ tests/lib/outputs/compliance/ -qGreen
ChangelogPR diffFragment prowler/changelog.d/<slug>.<type>.md, no direct CHANGELOG.md edit

Dashboard files (dashboard/compliance/<key with dots as underscores>.py) are optional: the local dashboard falls back to dashboard/compliance/generic.py. Do not request one.

Commands

# Structure + check existence (legacy and universal)
python3 skills/prowler-compliance-review/assets/validate_compliance.py prowler/compliance/<path>.json

# Load with the real model (prints framework, requirement count, providers)
uv run python -c "from prowler.lib.check.compliance_models import load_compliance_framework_universal as l; \
fw=l('prowler/compliance/<path>.json'); assert fw, 'see log'; print(fw.framework, len(fw.requirements), fw.get_providers())"

# Repo integrity tests for this file
uv run pytest tests/lib/check/compliance_catalog_integrity_test.py -k "<basename>" -q

# CLI: listed and requirements visible (no credentials needed)
uv run python prowler-cli.py <provider> --list-compliance | grep <key>
uv run python prowler-cli.py <provider> --list-compliance-requirements <key>

# Full run against a real account (writes output/compliance/)
uv run python prowler-cli.py <provider> --compliance <key> -M csv json-ocsf html

--compliance cannot be combined with --check, --service, --checks-file or --category (mutually exclusive argparse group).

Decision Tree

Loads with the real model?            No  -> FAIL
Integrity test / validator errors?    Yes -> FAIL (missing checks, duplicate ids, missing table_config...)
New framework authored as legacy?     Yes -> REQUEST CHANGES (use universal) unless extending a legacy family
Mappings padded or requirements missing? Yes -> REQUEST CHANGES
Changelog fragment missing?           Yes -> REQUEST CHANGES
Otherwise                             -> APPROVE

Common Issues

IssueDetectResolution
Typo in check idintegrity test / validatorFix id or remove it
Universal file without table_configvalidatorAdd outputs.table_config.group_by
Unknown attribute keyloader returns None, log says unknown attributeDeclare it in attributes_metadata or fix the key
Uppercase provider key in checksvalidatorUse prowler/providers/ directory names
Guardrail without Provider in universal filevalidatorAdd Provider
Automated requirement with no checks (CIS-style)validator warningAdd checks or mark Manual
Manual edit of CHANGELOG.mddiffMove to a changelog.d fragment

Resources

When not to use it

  • →Reviewing non-compliance related pull requests

Prerequisites

python3

Limitations

  • →Requires manual review of changelog fragments
  • →Dashboard files must follow a specific Python pattern

How it compares

It automates the validation of compliance framework structure and metadata, which would otherwise require manual verification against the codebase.

Compared to similar skills

prowler-compliance-review side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
prowler-compliance-review (this skill)05moReviewAdvanced
fix-dependabot-alerts189moReviewIntermediate
semgrep-rule-variant-creator15moReviewAdvanced
hooks-system19moReviewAdvanced

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

fix-dependabot-alerts

microsoft

Fix Dependabot security alerts by updating vulnerable npm dependencies. Use when the user mentions "dependabot", "security alerts", "vulnerability", "CVE", or wants to update packages with security issues.

1872

semgrep-rule-variant-creator

trailofbits

Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language.

10

hooks-system

MadAppGang

Comprehensive lifecycle hook patterns for Claude Code workflows. Use when configuring PreToolUse, PostToolUse, UserPromptSubmit, Stop, or SubagentStop hooks. Covers hook matchers, command hooks, prompt hooks, validation, metrics, auto-formatting, and security patterns. Trigger keywords - "hooks", "PreToolUse", "PostToolUse", "lifecycle", "tool matcher", "hook template", "auto-format", "security hook", "validation hook".

10

moai-workflow-loop

modu-ai

Ralph Engine - Automated feedback loop with LSP diagnostics and AST-grep integration for continuous code quality improvement. Use when implementing error-driven development, automated fixing, or continuous quality validation workflows.

10

yes

sstklen

Use when any task involves modifying files, configs, databases, or deployments. Use when debugging hits 2+ failures. Use when about to guess or assume without evidence ('probably', 'might be', 'I think', 'should be'). Use when deflecting to user ('please check...', 'you should manually...', 'you may

00

1password

openclaw

Set up and use 1Password CLI (op). Use when installing the CLI, enabling desktop app integration, signing in (single or multi-account), or reading/injecting/running secrets via op.

2799

Search skills

Search the agent skills registry