foto-webcam

What This Skill Does

The foto-webcam skill is a specialized agent capability designed for OpenClaw to interact with the foto-webcam.eu ecosystem. It allows users to manage a personal list of favorite webcam locations and retrieve real-time snapshots directly into the chat interface. The skill automates the process of identifying valid image URLs, fetching high-resolution snapshots from specified camera pages, and serving them as formatted attachments. By utilizing a local configuration file, users can maintain a structured repository of webcams, making it easy to monitor landscapes, weather, or city scenes across different locations.

Installation

To install this skill, use the ClawHub command-line interface. Ensure your environment has the necessary permissions to execute Python scripts within the designated workspace:

clawhub install openclaw/skills/skills/unixweb/foto-webcam

Use Cases

This skill is ideal for users who frequently check weather conditions or scenic views. Common scenarios include checking mountain weather before a trip, observing city traffic, or comparing multiple regional views simultaneously. The skill is designed to handle individual requests or batch requests (up to 6 images) to provide a comprehensive visual overview in one interaction. It acts as an bridge between static web pages and the chat interface, transforming raw URLs into usable visual data.

Example Prompts

1. webcam 1
2. webcam 1+2+3
3. liste

Tips & Limitations

• The skill relies on the file docs/webcams/favorites-muenchen.json for configuration. You can edit this file manually to add or remove favorites using the required JSON schema (id, name, page).
• When requesting multiple webcams, limit the request to a maximum of 6 images to ensure optimal performance and avoid API or network congestion.
• If a specific webcam page becomes unstable or changes its layout, you can manually update the entry in your favorites file by adding an 'image' field with a direct, static URL to the current JPG.
• Note that all text-based responses from the agent are provided in plain text without Markdown, ensuring compatibility with all audio and speech interfaces. The snapshot process is handled by a background Python script, which requires consistent network access to fetch images.