Connect OpenClaw to Dazzle to give your AI agent deep, personalized understanding of your life, memories, and preferences directly from your photos.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dazzle-photo-intelligence
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 dazzle-photo-intelligence 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 Dazzle Photo Intelligence?
Dazzle Photo Intelligence is a powerful bridge that connects OpenClaw to Dazzle, an AI agent that deeply understands the user through their photos. By leveraging a companion iOS app, Dazzle accesses and analyzes your photo library to build a rich understanding of who you are, where you live, your lifestyle preferences, and the key people in your life. This enables your AI assistant to deliver highly customized responses tailored to your real-world context rather than offering generic answers.
By integrating this bridge into your Openclaw Skills repository, you unlock an intelligent personalization layer. The skill operates as a local Model Context Protocol (MCP) server that is only invoked when personalized context is required. This guarantees that completely generic questions (such as coding queries or global facts) bypass the integration entirely, saving API cycles and ensuring data minimization.
Dazzle Photo Intelligence Use Cases
- Memories & Photo Search: Finding specific items, events, or trips (e.g., "show photos from last weekend" or "find vacation memories").
- Geographic Habits & Places: Querying restaurants, visited cities, or determining where the user currently resides.
- Tailored Recommendations: Requesting vacation planning suggestions or restaurant ideas based on previous likes.
- Personal Fact Summaries: Recalling family names, summarizing yearly activities, or tracking personal habits.
- Context-Aware Decisions: Asking the assistant to suggest appropriate gifts or plan meals tailored to specific tastes.
How Dazzle Photo Intelligence Works
- Context-Sensitive Tools: The bridge registers with OpenClaw and dynamically exposes capabilities. It presents fully functional Dazzle query tools if you are logged in, or fallback authentication tools if you are not.
- Secure User Authorization: A browser-based OAuth2 flow is triggered on the first tool call, avoiding static, long-lived API keys on your local storage.
- Local Encryption: Tokens are stored securely using an encrypted keyring setup powered by a user-generated password.
- Intelligent Query Resolution: Queries are routed through Dazzle to identify matching photo metadata, returning photo references within a structured JSON block.
- On-Demand Byte Fetching: Photo URLs are fetched directly only when the model needs to view, render, or analyze the actual image bytes.
Dazzle Photo Intelligence Setup
To deploy Dazzle Photo Intelligence within your Openclaw Skills suite, follow these setup steps to register it as an MCP server.
First, check if the bridge is already registered:
openclaw mcp show dazzle
If the server is not registered, perform the following step-by-step local configuration:
- Generate a secure, random password to encrypt your local sign-in state:
openssl rand -base64 24
- Register the MCP server by running the command below. Substitute
<KEYRING_PASSWORD>with your generated password and<SKILL_DIR>with the absolute path of this skill directory containingSKILL.md:
openclaw mcp set dazzle '{"command":"uvx","args":["--from","<SKILL_DIR>","--with","keyrings.alt","dazzle"],"env":{"DAZZLE_KEYRING_PASSWORD":"<KEYRING_PASSWORD>"}}'
- Verify the registration was successful:
openclaw mcp show dazzle
Dazzle Photo Intelligence Data Schema & Taxonomy
The skill manages access to personal files and metadata under a strict security schema. Data categorization is outlined below:
| Component | Type | Description |
|---|---|---|
| OAuth2 Tokens | Secure Storage | Short-lived, revocable access tokens mapped to the user's Dazzle account. |
| Keyring Passwords | Environment Variable | Configured via DAZZLE_KEYRING_PASSWORD in ~/.openclaw/openclaw.json. |
| Photo Results | <photos> JSON block |
Returned as a list of photo URLs to minimize payload sizes during initial queries. |
| User Context | Read-Only | Personal parameters such as inferred preferences, locations, and user-supplied descriptions. |
Queries are highly performant when scoped to narrow timeframes (e.g., a day or single trip). Year-long requests may require up to 60 seconds to execute, and tool timeouts should be adjusted accordingly.
Dazzle Photo Intelligence Advanced Features
- Dynamic State Adaptation: Only exposes operational query tools when authentication is active, preventing dead-ends or empty responses.
- Zero Static Keys: Employs industry-standard OAuth2, ensuring that long-lived static secrets are never committed or left unencrypted on disk.
- Complete Local Revocation: Running the uninstall routine clears local encryption credentials, making any cached token completely unrecoverable.
- On-Demand Visual Processing: Utilizes a reference-first approach where images are represented as URLs, saving processing overhead until multi-modal analysis is requested.
SKILL.md
Loading
Related Openclaw Skills
Facebook Page Ronnie
Automate your Facebook Page publishing and audience engagement via robust Graph API connections and intelligent browser fallback mechanisms.
bee-languages-get
A high-performance retrieval tool for obtaining a list of active site languages from the Bee Website Builder API to power dynamic localization workflows.
Google MediaPipe
Google MediaPipe is a cross-platform, on-device machine learning framework enabling highly optimized real-time vision, text, audio, and generative AI inference.
Image Watermark
An automated utility for processing and adding custom text watermarks to single or multiple images simultaneously.
Business Model Learner & Analyzer AI Agent
An AI-powered intellectual partner designed to help developers, entrepreneurs, and researchers systematically learn, analyze, and design robust business models.
A-Share Daily Brief
Automatically collects and generates beautifully structured daily briefs for the Chinese A-share stock market.
Featured*
