pageguard-mcp
Official# pageguard-mcp
MCP (Model Context Protocol) server that exposes PageGuard privacy compliance scanning as tools for AI coding assistants. Works with Claude Code, Cursor, Windsurf, ChatGPT, and any MCP-compatible environment.
## What it does
- **Local scan** — Detects tracking technologies, cookies, and third-party data collection from your project's `package.json`, config files, and `.env` files. No API key needed, no network requests.
- **URL scan** — Scans a live website for privacy compliance issues including risk scoring and compliance gap analysis.
- **Document generation** — Generates AI-written legal documents (privacy policy, terms of service, cookie policy, etc.) tailored to your detected technologies.
## Installation
### Claude Code
Add to your project's `.mcp.json` or global MCP config:
```json
{
"mcpServers": {
"pageguard": {
"command": "npx",
"args": ["pageguard-mcp"]
}
}
}
```
### Cursor
Add to Cursor Settings > MCP Servers:
```json
{
"mcpServers": {
"pageguard": {
"command": "npx",
"args": ["pageguard-mcp"]
}
}
}
```
### Windsurf
Add to your MCP configuration:
```json
{
"mcpServers": {
"pageguard": {
"command": "npx",
"args": ["pageguard-mcp"]
}
}
}
```
## Environment variables
| Variable | Required | Description |
|---|---|---|
| `PAGEGUARD_API_KEY` | No (local scan) / Yes (URL scan, doc gen) | API key from [getpageguard.com](https://getpageguard.com/#pricing) |
| `PAGEGUARD_API_URL` | No | Override API base URL (default: `https://www.getpageguard.com`) |
## Tools
### `pageguard_scan_local`
Scan a local project directory for privacy-relevant technologies.
**Input:**
- `path` (optional) — Absolute path to project directory. Defaults to current working directory.
**Output:** ComplianceReport JSON with detected technologies, data types, cookies, and third-party processors.
### `pageguard_scan_url`
Scan a live website URL for privacy compliance issues.
**Input:**
- `url` (required) — Full URL to scan, e.g. `https://example.com`
**Output:** ComplianceReport JSON with risk score, detected technologies, and compliance gaps.
### `pageguard_generate_docs`
Generate AI-written legal compliance documents for a scanned site.
**Input:**
- `scanId` (required) — Scan ID from a prior `pageguard_scan_url` result
- `documentType` (optional) — One of: `single` ($29), `bundle` ($49), `addon_security` ($19), `addon_a11y` ($19), `addon_schema` ($19), `app_bundle` ($39), `submission_guide` ($19). Defaults to `bundle`.
**Output:** Generated document content.
## Pricing
Scanning is free. Document generation requires credits:
- **Privacy Docs** ($29) — Privacy Policy + Terms of Service + Cookie Policy
- **Fix Everything** ($49) — All docs + Security Guide + Accessibility Report + Schema Markup
- **App Bundle** ($39) — Privacy docs + App Store Submission Guide
- **Add-ons** ($19 each) — Security Guide, Accessibility Report, Schema Markup, Submission Guide
- **Bulk packs** — 5 for $79, 15 for $149, 50 for $349
Get an API key at [getpageguard.com/#pricing](https://getpageguard.com/#pricing).
## License
MIT
TDQS
Scored across 3 tools
The three tools serve distinctly different purposes: scan_local analyzes codebase dependencies, scan_url analyzes live websites, and generate_docs produces legal documents from scan results. No functional overlap exists between them.
All tools follow a consistent pageguard_verb_noun snake_case pattern. The scanning tools use parallel naming (scan_local, scan_url) to distinguish targets, while generate_docs clearly indicates its document creation purpose.
Three tools is appropriate for this focused domain, covering the essential scan-local, scan-production, and generate-documentation workflow. While functional, the surface is minimal and could benefit from supporting tools like get_scan or list_documents.
The core workflow is covered: dual scanning capabilities and document generation. Minor gaps exist in document lifecycle management (no retrieval of previous scans or generated documents), but agents can work with immediate return values.