Fetch Kit
No description available
Ask AI about Fetch Kit
Powered by Claude Β· Grounded in docs
I know everything about Fetch Kit. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
Fetch Kit
A Go-based MCP (Model Control Protocol) connector that enables AI assistants like Claude to interact with web services. This tool provides a seamless interface for AI models to perform various web-related operations.
Features
- Retrieve web content using Jina AI
- Leverage Google AI (Gemini) services
- Support for various web service integrations
- Configurable tool groups
- Interact with web services through AI models
Installation
Requirements: Go 1.23.2+ (for building from source)
There are several ways to install Fetch Kit:
Option 1: Download from GitHub Releases
- Visit the GitHub Releases page
- Download the binary for your platform:
fetch-kit_linux_amd64for Linuxfetch-kit_darwin_amd64for macOSfetch-kit_windows_amd64.exefor Windows
- Make the binary executable (Linux/macOS):
chmod +x fetch-kit_* - Move it to your PATH (Linux/macOS):
sudo mv fetch-kit_* /usr/local/bin/fetch-kit
Option 2: Go Install
go install github.com/nguyenvanduocit/fetch-kit@latest
Option 3: Docker
Using Docker directly
- Pull the pre-built image from GitHub Container Registry:
docker pull ghcr.io/nguyenvanduocit/fetch-kit:latest - Or build the Docker image locally:
docker build -t fetch-kit .
Configuration
Environment Variables
The following environment variables are used for configuration:
# Required for AI services
GOOGLE_AI_API_KEY= # Required: API key for Google AI (Gemini) service
JINA_API_KEY= # Required: API key for Jina AI service
# Optional configurations
ENABLE_TOOLS= # Optional: Comma-separated list of tool groups to enable (empty = all enabled)
PROXY_URL= # Optional: HTTP/HTTPS proxy URL if needed
You can set these:
- Directly in the Docker run command (recommended)
- Through a
.envfile (use the-envflag) - Directly in your shell environment
Using with Claude and Cursor
To make Fetch Kit work with Claude and Cursor, you need to add configuration to your Cursor settings.
Step 1: Install Fetch Kit
Choose one of the installation methods above (Docker recommended).
Step 2: Configure Cursor
- Open Cursor
- Go to Settings > MCP > Add MCP Server
- Add the following configuration:
Option A: Using Docker (Recommended)
{
"mcpServers": {
"fetch_kit": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "GOOGLE_AI_API_KEY=your_google_ai_key",
"-e", "JINA_API_KEY=your_jina_api_key",
"ghcr.io/nguyenvanduocit/fetch-kit:latest"
]
}
}
}
Option B: Using Local Binary
{
"mcpServers": {
"fetch_kit": {
"command": "fetch-kit",
"args": ["-env", "/path/to/.env"]
}
}
}
Step 3: Test Connection
You can test if the connection is working by asking Claude in Cursor:
@fetch_kit get web content from https://example.com
Enable Tools
Fetch Kit supports various tool groups that can be enabled or disabled using the ENABLE_TOOLS environment variable. If not specified, all tools are enabled by default.
CLI Usage
In addition to the MCP server, fetch-kit ships a standalone CLI binary (fetch-cli) for direct terminal use β no MCP client needed.
Installation
# Build locally
just install-cli
# Or with Go
go install github.com/nguyenvanduocit/fetch-kit/cmd/fetch-cli@latest
Quick Start
# Set credentials (only needed for the tools you use)
export JINA_API_KEY=your-jina-key
export GOOGLE_AI_API_KEY=your-gemini-key
# Or use a .env file
fetch-cli --env .env <command> [flags]
Commands
| Command | Description |
|---|---|
jina-fetch | Fetch and extract web content via Jina AI |
gemini-fetch | Fetch and analyze content using Gemini AI |
youtube | Get YouTube video transcripts |
Examples
# Fetch a web page
fetch-cli jina-fetch --url https://example.com
# Ask Gemini about a URL
fetch-cli gemini-fetch --question "Summarize this" --context "https://example.com"
# Get YouTube transcript
fetch-cli youtube --video-id dQw4w9WgXcQ
# JSON output
fetch-cli jina-fetch --url https://example.com --output json
Flags
Every command accepts:
--env stringβ Path to.envfile (default:.env)--output stringβ Output format:text(default) orjson
Contributing
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
For a list of recent changes, see CHANGELOG.md.
Installation
Homebrew (macOS/Linux)
brew install nguyenvanduocit/tap/fetch-kit
