Baoyu Gemini Web
Access and analyze live web data using Gemini's power, guided by Jim Liu’s specialized BaoYu framework.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add baoyu-gemini-web npx -- -y @trustedskills/baoyu-gemini-web
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"baoyu-gemini-web": {
"command": "npx",
"args": [
"-y",
"@trustedskills/baoyu-gemini-web"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
This skill, Baoyu Gemini Web, allows AI agents to access and analyze live web data using Google's Gemini model through a reverse-engineered API. It supports text generation, image generation (with download and saving capabilities), and the ability to provide reference images for vision input. The skill also facilitates multi-turn conversations by persisting session IDs. It is built upon Jim Liu’s BaoYu framework and requires an initial consent check before use.
When to use it
- Generating text content based on current web trends or information.
- Creating images based on prompts, with the ability to save them locally.
- Analyzing visual data by providing reference images alongside textual instructions.
- Maintaining context across multiple interactions within a conversation.
- When you need access to Gemini's capabilities but cannot use the official Google API.
Key capabilities
- Text generation using Gemini.
- Image generation and local saving.
- Vision input with reference images.
- Multi-turn conversations via session persistence (
--sessionId). - Reverse-engineered Gemini Web API access (not an official Google API).
- Consent check flow to acknowledge risks associated with the reverse-engineered API.
Example prompts
- "Generate a short story about [topic] based on current news."
- "Create an image of [description] and save it as 'my_image.png'."
- "Analyze this image [attach image file] and describe what's happening."
- “Continue our conversation from the last time, I was asking about…”
Tips & gotchas
- Consent is required: Before using this skill, you must perform a consent check as outlined in the documentation. This involves verifying an existing consent file or acknowledging a disclaimer and accepting terms.
- Reverse-engineered API: The skill uses a reverse-engineered Gemini Web API, which means it's not officially supported by Google and may break if the underlying API changes.
- Script location: All scripts for this skill are located in the
scripts/subdirectory within the skill’s directory.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.