A zero-token automation pipeline that identifies faces in WhatsApp media and backs them up to Google Photos via GCP.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dmaf
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install dmaf using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is DMAF (Don't Miss A Face)?
DMAF (Don't Miss A Face) is a sophisticated media management tool that automates the transition of photos and videos from WhatsApp groups to Google Photos. By leveraging face recognition technology, it filters incoming media to ensure only the people you care about are archived.
This skill is a standout among Openclaw Skills because it is designed for maximum efficiency. While an AI agent handles the complex initial setup on Google Cloud Platform (GCP), the day-to-day operations run on a system cron and Cloud Run. This architecture ensures that once the pipeline is live, it consumes zero LLM tokens, making it a cost-effective solution for long-term media preservation.
DMAF (Don't Miss A Face) Use Cases
- Automatically preserve family photos shared in high-volume WhatsApp group chats.
- Organize event photography by identifying and sorting specific participants into Google Photos.
- Create a hands-free backup system for personal media without manual downloading.
- Monitor multiple groups simultaneously and consolidate relevant media into a single searchable cloud library.
How DMAF (Don't Miss A Face) Works
- OpenClaw automatically downloads incoming media from targeted WhatsApp groups to a local directory.
- A synchronization script moves files from the local directory to a Google Cloud Storage (GCS) staging bucket.
- Cloud Scheduler triggers a Cloud Run Job on an hourly basis to process new media.
- The system compares faces in the media against a reference database of known people stored in GCS.
- Matches are automatically uploaded to Google Photos, while Firestore tracks file hashes to prevent any duplicate processing.
- For videos, the system samples frames at 1-2fps and performs an early exit upon the first face match to save compute resources.
DMAF (Don't Miss A Face) Setup
To begin using this skill from the Openclaw Skills collection, follow these steps:
# Install the DMAF skill using ClaWHub
clawhub install dmaf
# Or manually copy the skill to your local directory
cp -r deploy/openclaw-skill/ ~/.openclaw/skills/dmaf/
Once installed, provide your GCP Project ID to the agent. The agent will guide you through enabling the necessary GCP APIs, setting up service accounts, creating GCS buckets, and configuring the Google Photos OAuth tokens within Secret Manager.
DMAF (Don't Miss A Face) Data Schema & Taxonomy
The skill manages data across several GCP services to ensure reliability and speed:
| Component | Purpose | Data Type |
|---|---|---|
| GCS Bucket (Media) | Staging area for incoming WhatsApp files | Binary Media |
| GCS Bucket (People) | Reference images for recognition (one folder per person) | JPG/PNG |
| Firestore | Deduplication database using SHA-256 hashes | NoSQL Documents |
| Secret Manager | Stores configuration YAML and OAuth tokens | Encrypted Strings |
| Cloud Logging | Stores execution logs for auditing and troubleshooting | Structured Logs |
DMAF (Don't Miss A Face) Advanced Features
- Video Recognition: Analyzes video clips and uploads the full file upon a positive face match.
- Tokenless Operation: Operates entirely on native cloud infrastructure after setup, requiring no ongoing AI API calls.
- MCP Server Integration: Ships with an MCP server for use with Claude Desktop, Claude Code, and Cursor.
- Automated Deduplication: Ensures the same file is never processed or uploaded twice via Firestore indexing.
- Scalable Reference Library: Easily add new people by simply creating a new subdirectory in the reference GCS bucket.
SKILL.md
Loading
Related Openclaw Skills
ClawTourism
An automated travel intelligence skill that converts Gmail booking confirmations into structured itineraries with proactive alerts and monitoring.
ClawSavings — Israeli Discounts & Price Comparison
An AI agent skill that identifies the best Israeli credit card clubs, vouchers, and store discounts to maximize consumer savings.
ClawEvents
A multi-city event discovery engine for finding concerts, cinema, and local happenings in Tel Aviv, Barcelona, and New York.
Clawback
Manage group expenses and bill splitting through natural language commands within WhatsApp or Telegram groups.
GClaw
A sophisticated Gmail intelligence tool for Openclaw Skills that enables heuristic email classification and automated digestion without LLM overhead.
ClawCierge
A global restaurant search and booking skill for AI agents with integrated Michelin and Time Out recommendations.
Featured*
