whatsapp-import

whatsapp-import Agent

Import contacts and interactions from WhatsApp .txt exports.

Engram Injection

Before starting work, load relevant learned patterns:

1. Preferred: Call plur_admin MCP tool with action = "plur_inject_hybrid", prompt = your task description, scope = agent:whatsapp-import
2. Fallback: If MCP is unavailable, read .datacore/state/agent-engrams/whatsapp-import.md for compiled engrams

Engrams encode learned behavioral patterns that improve task quality.

Purpose

Process WhatsApp chat exports to:

1. Create CRM contacts from chat participants
2. Extract interaction history for CRM adapter
3. Build phone-to-contact mapping index

Trigger

• Manual via /whatsapp import
• When new files appear in .datacore/state/whatsapp/exports/

Workflow

1. Discover Exports

from pathlib import Path
from lib import WhatsAppExportParser, parse_export_directory

exports_dir = Path.home() / 'Data' / '.datacore' / 'state' / 'whatsapp' / 'exports'
exports = parse_export_directory(exports_dir)

2. Preview Import

Before creating contacts, show preview:

from lib import WhatsAppContactCreator

creator = WhatsAppContactCreator()
preview = creator.get_import_preview()

print(f"Exports: {preview['export_count']}")
print(f"Contacts to create: {preview['candidate_count']}")

for c in preview['candidates'][:10]:
    print(f"  {c['name']}: {c['message_count']} messages")

3. Create Contacts

results = creator.create_contacts_from_exports(
    space='0-personal',
    dry_run=False
)

print(f"Created: {len(results['created'])}")
print(f"Matched existing: {len(results['matched'])}")
print(f"Skipped (low activity): {len(results['skipped'])}")

4. Post-Processing

After successful import:

1. Move processed exports to processed/ subdirectory
2. Update phone-index.yaml with new mappings
3. Report results to user

Input

• Export files in .datacore/state/whatsapp/exports/*.txt
• Optional: specific file path to import