Skip to main content
Viby Logo

Viby

Viby vibes everything

GitHub Repo PyPI version Python Version License: GPL v3 UV MCP Ask DeepWiki

English | 中文

🚀 Overview

Viby is a powerful AI agent that lives in your terminal, designed to solve virtually any task you throw at it. Whether you need code assistance, shell commands, information retrieval, or creative content - Viby vibes with your needs and delivers solutions instantly.

✨ Features

  • Intelligent Conversations - Engage in natural multi-turn dialogues
  • Automatic Shell Command Generation - Get optimized shell commands when needed
  • Pipeline Integration - Process data from other commands (e.g., git diff | viby "write a commit message")
  • MCP Tools - Extended capabilities through Model Context Protocol integration
  • Smart Tool Discovery - Automatically finds and uses the most relevant tools within configured MCP servers
  • Enhanced History Management - Complete interaction history with search, export, and management
  • Multiple Model Support - Configure and use different models for various tasks
  • Command Embeddings - Semantic search in tools using embedded vectors for accurate tool selection
  • Multi-language Support - Full interface in English and Chinese with easy language switching

🔧 Installation

# Install from PyPI
pip install viby
# OR
uv tool install viby

Alternative Installation

# Install from source with uv
uv pip install -e .

Usage Examples

Basic Question

yb vibe "Write a quicksort in python"
# -> Sure! Here is a quicksort algorithm implemented in **Python**:

Simplified Command

yb "Write a quicksort in python"
# -> Same result as above

Process Piped Content

git diff | yb vibe "Generate a commit message"
# -> Added information to the README
yb vibe "What is this project about?" < README.md
# -> This project is about...

Generate Shell Command

yb vibe "How many lines of python code did I write?"
# -> find . -type f -name "*.py" | xargs wc -l
# -> [r]run, [e]edit, [y]copy, [c]chat, [q]quit (default: run): 

Advanced Model Selection

# Use think model for complex analysis
yb --think vibe "Analyze this complex algorithm and suggest optimizations"

# Use fast model for quick responses
yb --fast vibe "Translate 'Hello, World!' to French"

Shell Command Magic Integration

# List directory contents
yb vibe "$(ls) What files are in the current directory?"
# -> The current directory contains: file1.txt, file2.py, directory1/...

# Analyze Git status
yb vibe "$(git status) Which files should I commit first?"

# View code files
yb vibe "$(cat main.py) How can I improve this code?"

Smart Tool Discovery

# Viby will automatically discover and use relevant tools
yb vibe "What's the weather in San Francisco?"
# -> [Viby identifies and uses weather tools]
# -> The current weather in San Francisco is 68°F and partly cloudy...

# Embedding Model Management
# First download the embedding model (required once before using embedding features)
# Embed model configurable with yb --config
yb tools embed download

# Start the embedding server (required for tool discovery)
yb tools embed start

# Check embedding server status
yb tools embed status

# Update tool embeddings from configured MCP servers
yb tools embed update

# List available tools (udpate before listing)
yb tools list

# Stop the embedding server when not needed
yb tools embed stop

History Management

# View recent interactions
yb history list

# Search your history
yb history search "python"

# Export your interaction history
yb history export history.json

# View shell command history
yb history shell

# Clear history (with confirmation)
yb history clear

Automatically Use MCP Tools When Needed

yb vibe "What time is it now?"
# -> [AI uses time tool to get current time]
# -> "datetime": "2025-05-03T00:49:57+08:00"

Keyboard Shortcuts

Viby provides a convenient keyboard shortcut (Ctrl+Q) that allows you to quickly use Viby with the current command line content:

# Install the keyboard shortcuts (auto-detects your shell)
yb shortcuts

# After installation, type any command and press Ctrl+Q
help me analyze my readme file  # Now press Ctrl+Q
# -> This transforms into: yb vibe help me analyze my readme file
# -> [AI analyzes and responds to question]

Supported shells:

  • Bash
  • Zsh
  • Fish

After installing shortcuts, you'll need to reload your shell configuration (source ~/.bashrc, source ~/.zshrc, or equivalent) or restart your terminal for the shortcuts to take effect.

Command Structure

Viby uses a simple command structure:

yb [OPTIONS] [COMMAND] [ARGS]...

Main commands:

  • yb [prompt] - Ask a question (alias for yb vibe "your question"; recommended)
  • yb vibe "your question" - Ask a question (default command for questions)
  • yb --think vibe "complex question" - Use the think model for deeper analysis
  • yb --fast vibe "simple question" - Use the fast model for quick responses
  • yb history - Manage interaction history
  • yb tools - Manage tool-related commands
  • yb shortcuts - Install keyboard shortcuts

Use yb --help to see all available commands and options.

Configuration

Viby reads configuration from ~/.config/viby/config.yaml. You can set the model, parameters, and MCP options here.

Interactive Configuration

Use the configuration wizard to set up your preferences:

yb --config

This allows you to configure:

  • API endpoint and key
  • Model
  • Temperature and token settings
  • MCP tools enablement
  • Interface language
  • Embedding model settings

MCP Server Configuration

Viby supports Model Context Protocol (MCP) servers for extended capabilities. MCP configurations are stored in ~/.config/viby/mcp_servers.json.

⭐ Star History

📄 Documentation

🤝 Contributing

Contributions are welcome! Feel free to submit a Pull Request or create an Issue.

Metadata

Release files for viby 0.2.5

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for viby 0.2.5
File Size Uploaded
viby-0.2.5.tar.gz 89.1 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for viby 0.2.5
File Interpreter ABI Platform
viby-0.2.5-py3-none-any.whl Python 3 none any Details

Total release size: 192.2 kB

Release files / viby-0.2.5.tar.gz

Download URL viby-0.2.5.tar.gz
Size 89.1 kB
Tags Source
SHA-256 checksum
How to use checksums
4cbe5322371c8e8308e5c9a7a6632e9258e5e51a084bde3337d33d073af2c4c3
BLAKE2b-256 checksum
How to use checksums
3c565bd1c3ac44f2e3587c7c65b2ab5dfaf002fec9f1e2cdb72ff5aa82a36594
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.12.9

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on May 17, 2025.

Transparency log

Release files / viby-0.2.5-py3-none-any.whl

Download URL viby-0.2.5-py3-none-any.whl
Size 103.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
df99db790c83d67fa32ee1a85023a1a62863452f196849a5845d904cef95d9ea
BLAKE2b-256 checksum
How to use checksums
db5bb646fa8c62b72674e596a35a36581981338d59bbc7e915ba68ea6be98d7a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.12.9

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on May 17, 2025.

Transparency log
Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page