Agent Skills: Notion Data Handling

Notion Data Handling

Overview

Handle sensitive data correctly when integrating with Notion: detect PII in page properties and block content, redact sensitive fields before logging or exporting, minimize data exposure with filter_properties, and implement GDPR/CCPA compliance patterns including right-of-access exports, right-of-deletion (archive or field clearing), and retention-based archival with audit logging.

Prerequisites

• @notionhq/client v2+ installed (npm install @notionhq/client)
• Python alternative: notion-client (pip install notion-client)
• Understanding of which Notion databases contain personal data
• Audit logging infrastructure (structured logs, SIEM, or Notion audit database)
• Legal guidance on applicable regulations (GDPR, CCPA, HIPAA, etc.)

Instructions

Step 1: PII Detection in Notion Content

Notion pages can contain PII in any property type. Scan systematically:

import { Client } from '@notionhq/client';
import type { PageObjectResponse } from '@notionhq/client/build/src/api-endpoints';

const notion = new Client({ auth: process.env.NOTION_TOKEN });

// PII pattern matchers
const PII_PATTERNS = [
  { type: 'email',      pattern: /[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}/g },
  { type: 'phone_us',   pattern: /\b\d{3}[-.]?\d{3}[-.]?\d{4}\b/g },
  { type: 'phone_intl', pattern: /\+\d{1,3}[-.\s]?\d{4,14}/g },
  { type: 'ssn',        pattern: /\b\d{3}-\d{2}-\d{4}\b/g },
  { type: 'credit_card', pattern: /\b\d{4}[-\s]?\d{4}[-\s]?\d{4}[-\s]?\d{4}\b/g },
  { type: 'ip_address', pattern: /\b\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\b/g },
];

interface PIIFinding {
  propertyName: string;
  piiType: string;
  location: 'property' | 'content';
}

function scanPageForPII(page: PageObjectResponse): PIIFinding[] {
  const findings: PIIFinding[] = [];

  for (const [name, prop] of Object.entries(page.properties)) {
    // Direct PII property types
    if (prop.type === 'email' && prop.email) {
      findings.push({ propertyName: name, piiType: 'email', location: 'property' });
    }
    if (prop.type === 'phone_number' && prop.phone_number) {
      findings.push({ propertyName: name, piiType: 'phone', location: 'property' });
    }
    if (prop.type === 'people' && prop.people.length > 0) {
      findings.push({ propertyName: name, piiType: 'user_reference', location: 'property' });
    }

    // Text properties may contain embedded PII
    if (prop.type === 'rich_text' || prop.type === 'title') {
      const textParts = prop.type === 'title' ? prop.title : prop.rich_text;
      const text = textParts.map(t => t.plain_text).join('');

      for (const { type, pattern } of PII_PATTERNS) {
        // Reset regex lastIndex for each check
        pattern.lastIndex = 0;
        if (pattern.test(text)) {
          findings.push({ propertyName: name, piiType: type, location: 'property' });
        }
      }
    }
  }

  return findings;
}

// Scan an entire database for PII
async function auditDatabaseForPII(dbId: string) {
  const findings: { pageId: string; pageTitle: string; pii: PIIFinding[] }[] = [];
  let cursor: string | undefined;

  do {
    const response = await notion.databases.query({
      database_id: dbId,
      page_size: 100,
      start_cursor: cursor,
    });

    for (const page of response.results) {
      if (!('properties' in page)) continue;
      const pii = scanPageForPII(page as PageObjectResponse);
      if (pii.length > 0) {
        const titleProp = Object.values(page.properties)
          .find(p => p.type === 'title');
        const title = titleProp?.type === 'title'
          ? titleProp.title.map(t => t.plain_text).join('')
          : 'Untitled';
        findings.push({ pageId: page.id, pageTitle: title, pii });
      }
    }

    cursor = response.has_more ? response.next_cursor ?? undefined : undefined;
  } while (cursor);

  return findings;
}

Python — PII scanner:

import re
from notion_client import Client

client = Client(auth=os.environ["NOTION_TOKEN"])

PII_PATTERNS = [
    ("email", re.compile(r"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}")),
    ("phone", re.compile(r"\b\d{3}[-.]?\d{3}[-.]?\d{4}\b")),
    ("ssn", re.compile(r"\b\d{3}-\d{2}-\d{4}\b")),
]

def scan_page_for_pii(page: dict) -> list[dict]:
    findings = []
    for name, prop in page["properties"].items():
        if prop["type"] == "email" and prop.get("email"):
            findings.append({"property": name, "type": "email"})
        if prop["type"] == "phone_number" and prop.get("phone_number"):
            findings.append({"property": name, "type": "phone"})
        if prop["type"] in ("rich_text", "title"):
            parts = prop.get("title" if prop["type"] == "title" else "rich_text", [])
            text = "".join(t["plain_text"] for t in parts)
            for pii_type, pattern in PII_PATTERNS:
                if pattern.search(text):
                    findings.append({"property": name, "type": pii_type})
    return findings

Step 2: Redaction and Data Minimization

Redact PII before logging or exporting:

function redactPageProperties(
  page: PageObjectResponse,
  sensitiveFields: string[] = ['Email', 'Phone', 'SSN']
): Record<string, unknown> {
  const redacted: Record<string, unknown> = { id: page.id };

  for (const [name, prop] of Object.entries(page.properties)) {
    // Always redact known sensitive property types
    if (prop.type === 'email') {
      redacted[name] = prop.email ? '[REDACTED_EMAIL]' : null;
      continue;
    }
    if (prop.type === 'phone_number') {
      redacted[name] = prop.phone_number ? '[REDACTED_PHONE]' : null;
      continue;
    }
    if (prop.type === 'people') {
      redacted[name] = `[${prop.people.length} users]`;
      continue;
    }

    // Redact explicitly marked sensitive fields
    if (sensitiveFields.includes(name)) {
      redacted[name] = '[REDACTED]';
      continue;
    }

    // Safe property types pass through
    switch (prop.type) {
      case 'title':
        redacted[name] = prop.title.map(t => t.plain_text).join('');
        break;
      case 'select':
        redacted[name] = prop.select?.name ?? null;
        break;
      case 'multi_select':
        redacted[name] = prop.multi_select.map(s => s.name);
        break;
      case 'number':
        redacted[name] = prop.number;
        break;
      case 'checkbox':
        redacted[name] = prop.checkbox;
        break;
      case 'date':
        redacted[name] = prop.date?.start ?? null;
        break;
      default:
        redacted[name] = `[${prop.type}]`;
    }
  }

  return redacted;
}

// Safe logging — never log raw page objects
console.log('Processing page:', JSON.stringify(redactPageProperties(page)));
// NEVER: console.log('Page:', JSON.stringify(page)); // LEAKS PII

Data minimization — only request properties you need:

// filter_properties limits which properties are returned by the API
async function getTaskStatuses(dbId: string) {
  const response = await notion.databases.query({
    database_id: dbId,
    filter_properties: ['Status', 'Name', 'Due Date'],
    page_size: 100,
  });
  // Response only contains Status, Name, Due Date — no email, phone, etc.
  return response;
}

Step 3: GDPR/CCPA Compliance Patterns

Right of Access — export all data for a user:

async function exportUserData(userId: string, databaseIds: string[]) {
  const exportData: Record<string, unknown> = {
    exportedAt: new Date().toISOString(),
    requestType: 'GDPR Article 15 — Right of Access',
    source: 'Notion Integration',
    databases: {} as Record<string, unknown>,
  };

  for (const dbId of databaseIds) {
    const response = await notion.databases.query({
      database_id: dbId,
      filter: {
        property: 'Assignee',
        people: { contains: userId },
      },
    });

    (exportData.databases as Record<string, unknown>)[dbId] = response.results
      .filter((p): p is PageObjectResponse => 'properties' in p)
      .map(page => ({
        id: page.id,
        url: page.url,
        created: page.created_time,
        lastEdited: page.last_edited_time,
        properties: page.properties,
      }));
  }

  // Audit log the export
  console.log(JSON.stringify({
    event: 'gdpr_data_export',
    userId,
    databaseCount: databaseIds.length,
    timestamp: new Date().toISOString(),
  }));

  return exportData;
}

Right of Deletion — archive pages or clear PII fields:

async function deleteUserData(
  userId: string,
  databaseIds: string[],
  strategy: 'archive' | 'clear_pii' = 'archive'
) {
  const deletionLog: { pageId: string; action: string; database: string }[] = [];

  for (const dbId of databaseIds) {
    const pages = await notion.databases.query({
      database_id: dbId,
      filter: {
        property: 'Assignee',
        people: { contains: userId },
      },
    });

    for (const page of pages.results) {
      if (strategy === 'archive') {
        // Soft delete — page moved to trash (recoverable for 30 days)
        await notion.pages.update({
          page_id: page.id,
          archived: true,
        });
        deletionLog.push({ pageId: page.id, action: 'archived', database: dbId });
      } else {
        // Clear PII fields but keep the record
        await notion.pages.update({
          page_id: page.id,
          properties: {
            Email: { email: null },
            Phone: { phone_number: null },
            Assignee: { people: [] },
            Notes: { rich_text: [{ text: { content: '[Data deleted per GDPR request]' } }] },
          },
        });
        deletionLog.push({ pageId: page.id, action: 'pii_cleared', database: dbId });
      }

      // Rate limit: 3 requests/second
      if (deletionLog.length % 3 === 0) {
        await new Promise(r => setTimeout(r, 1100));
      }
    }
  }

  // Audit log (REQUIRED for compliance — keep for minimum retention period)
  console.log(JSON.stringify({
    event: 'gdpr_data_deletion',
    userId,
    strategy,
    pagesAffected: deletionLog.length,
    timestamp: new Date().toISOString(),
    log: deletionLog,
  }));

  return deletionLog;
}

Data retention — archive pages past retention window:

async function enforceRetention(dbId: string, retentionDays: number) {
  const cutoff = new Date();
  cutoff.setDate(cutoff.getDate() - retentionDays);

  let cursor: string | undefined;
  let archived = 0;

  do {
    const response = await notion.databases.query({
      database_id: dbId,
      filter: {
        timestamp: 'last_edited_time',
        last_edited_time: { before: cutoff.toISOString() },
      },
      page_size: 100,
      start_cursor: cursor,
    });

    for (const page of response.results) {
      await notion.pages.update({ page_id: page.id, archived: true });
      archived++;
      // Respect rate limits
      if (archived % 3 === 0) await new Promise(r => setTimeout(r, 1100));
    }

    cursor = response.has_more ? response.next_cursor ?? undefined : undefined;
  } while (cursor);

  console.log(JSON.stringify({
    event: 'retention_enforcement',
    database_id: dbId,
    retention_days: retentionDays,
    pages_archived: archived,
    cutoff_date: cutoff.toISOString(),
    timestamp: new Date().toISOString(),
  }));

  return { archived, cutoffDate: cutoff.toISOString() };
}

Output

• PII detection scanning all property types and text content (TS + Python)
• Redaction layer preventing PII leakage in logs and exports
• Data minimization via filter_properties in API queries
• GDPR Article 15 data export with audit logging
• GDPR Article 17 deletion (archive or field clearing) with rate limiting
• Retention-based archival with structured compliance logging
• Audit trail for all data access, export, and deletion events

Error Handling

| Issue | Cause | Solution | |-------|-------|----------| | PII in application logs | Missing redaction layer | Use redactPageProperties for all logging | | Deletion fails on pages | Integration lacks Update capability | Edit integration at notion.so/my-integrations | | Export missing pages | Pagination not handled | Use start_cursor loop until has_more is false | | Rate limit during bulk deletion | Too many update calls | Throttle to 3 requests/second with delays | | Regex false positives | Overly broad patterns | Tune patterns for your data; consider allowlists | | Audit log gaps | Async logging dropped events | Use synchronous logging for compliance events |

Examples

Quick PII Audit for a Database

const findings = await auditDatabaseForPII(process.env.NOTION_DB_ID!);
console.log(`PII audit: ${findings.length} pages with PII detected`);
for (const f of findings) {
  console.log(`  Page "${f.pageTitle}": ${f.pii.map(p => p.piiType).join(', ')}`);
}

Python Data Export

def export_user_data(user_id: str, db_ids: list[str]) -> dict:
    export = {"exported_at": datetime.utcnow().isoformat(), "databases": {}}
    for db_id in db_ids:
        results = client.databases.query(
            database_id=db_id,
            filter={"property": "Assignee", "people": {"contains": user_id}},
        )
        export["databases"][db_id] = [
            {"id": p["id"], "properties": p["properties"]}
            for p in results["results"]
        ]
    return export

Resources

• GDPR Developer Guide — key articles for data processors
• Notion Page Properties Reference — all property types
• Database Query with filter_properties — data minimization
• CCPA Overview — California Consumer Privacy Act requirements
• Notion API Update Page — archive and property updates

Next Steps

For enterprise access control and multi-workspace permissions, see notion-enterprise-rbac.