compliance-framework

name	Compliance Framework
description	Technology-agnostic, referential-agnostic orchestration framework for all compliance standards (RGAA, RGESN, RGS, RGPD, RGI, W3C-WSG)

SKILL: Compliance Framework (Shared Infrastructure)

Note: This is not a traditional "skill" with tests or criteria. It's a shared framework used by all compliance referential skills to orchestrate scanning across different build systems and projects.

🎯 Purpose

Provides universal orchestration for compliance scanning:

• Technology-agnostic: Works with Nx, Turborepo, pnpm, Maven, Gradle, or custom projects
• Referential-agnostic: Serves all compliance standards (RGAA, RGESN, RGS, RGPD, RGI, W3C-WSG)
• Configuration-driven: Projects provide minimal config; framework handles the rest
• Agent-ready: Enables agents to invoke multiple skills based on project context

📦 What's Included

Core Infrastructure

• scripts/scan-affected.js — Universal wrapper

  • Auto-detects build system (Nx, Turborepo, etc.)
  • Finds affected projects via git/build commands
  • Maps projects to compliance targets
  • Invokes referential-specific scanners

• scripts/config-template.json — Template for projects

  • Define compliance targets (e.g., web app, component library)
  • Map projects to targets (e.g., web → web, api → null)
  • Used by ALL referential skills

Documentation

• ARCHITECTURE.md — Design pattern (tech + referential agnostic)
• README.md — Framework usage guide & all supported build systems
• AGENT-INTEGRATION.md — How agents select and invoke skills

🚀 Usage

By Referential Skills

Each compliance skill (RGAA, RGESN, etc.) can use the framework:

cd .agents/skills/compliance-rgaa
node ../compliance-framework/scripts/scan-affected.js

Framework auto-discovers the skill's scan-compliance.js and invokes it.

Explicit Scanner Path

node .agents/skills/compliance-framework/scripts/scan-affected.js \
  --scanner .agents/skills/compliance-rgesn/scripts/scan-compliance.js \
  --base main \
  --targets web

By Agents

Agents select relevant skills and invoke the framework for each:

@compliance-agent Review for French compliance
  ↓
Agent selects: RGAA, RGPD, RGS, RGESN, RGI
  ↓
For each skill:
  node ... --scanner .agents/skills/compliance-[skill]/scripts/scan-compliance.js

🔧 Supported Technologies

Technology	Detection	Affected Projects	Fallback
Nx	nx.json	nx show projects --affected	git diff
Turborepo	turbo.json	git diff	—
pnpm	pnpm-workspace.yaml	git diff	—
Maven	pom.xml	git diff	—
Gradle	build.gradle(*.kts)	git diff	—
Custom	None detected	git diff or fallback	—

📋 All Compliance Skills Using This Framework

Skill	Location	Standard	Tests
RGAA	compliance-rgaa/	Accessibility	182
RGESN	compliance-rgesn/	Sustainability	TBD
RGS	compliance-rgs/	Security	TBD
RGPD	compliance-rgpd/	Data Protection	TBD
RGI	compliance-rgi/	Interoperability	TBD
W3C-WSG	compliance-w3c-wsg/	Web Sustainability	TBD

✨ Key Features

✅ Zero duplication — Framework shared by all skills
✅ Minimal project config — One scanner-config.json for all standards
✅ Automatic build detection — No manual setup per project
✅ Agent-driven — Agents select applicable referentials
✅ Extensible — Add new technologies with 5 lines of code
✅ Well-documented — See README.md and ARCHITECTURE.md for details

📖 Learn More

• ARCHITECTURE.md — Design pattern & responsibility separation
• README.md — Usage guide & configuration
• AGENT-INTEGRATION.md — Agent selection logic

---

Type: Shared Framework (Standard: VS Code .agents/skills/)
Applicable To: All compliance referential skills & projects
Created: March 2026