Click on the screenshot to watch the full demonstration.

Prerequisites

Anthropic API Key

Opsy uses Anthropic's Claude AI models to provide intelligent assistance. You'll need an Anthropic API key:

1. Create an account at Anthropic's website

2. Generate an API key from your account dashboard

3. Set the API key in your Opsy configuration (see Configuration section) or as an environment variable:

   export ANTHROPIC_API_KEY=your_api_key_here
   

Command-Line Tools

Opsy works with standard command-line tools. While none are strictly required to run Opsy, having them installed expands its capabilities:

• Git - Version control
• GitHub CLI - GitHub integration
• kubectl - Kubernetes management
• AWS CLI - AWS management
• Helm - Kubernetes package manager
• Google Cloud CLI (gcloud) - Google Cloud management
• Jira CLI - Jira automation

Opsy adapts to your environment and only uses tools that are installed on your system.

Installation

Via Go Install

For users with Go 1.24 or later:

go install github.com/datolabs-io/opsy/cmd/opsy@latest

Ensure your Go bin directory is in your PATH.

Via Homebrew

For macOS and Linux users with Homebrew:

brew tap datolabs-io/opsy https://github.com/datolabs-io/opsy
brew install datolabs-io/opsy/opsy

Direct Download

Each release includes binaries for various platforms:

1. Download the appropriate binary for your operating system
2. Make it executable (Unix-based systems): chmod +x opsy
3. Move it to a directory in your PATH: mv opsy /usr/local/bin/ (or another directory in your PATH)

Usage

Opsy is simple to use. Just describe what you want to do in plain language, and Opsy will handle the rest.

opsy 'Your task description here'

For example:

# Repository management
opsy 'Create a new private repository in datolabs-io organization named backup'

# Kubernetes troubleshooting
opsy 'Check why pods in the production namespace are crashing'

# Log analysis
opsy 'Find errors in the application logs from the last hour'

Opsy interprets your instructions, builds a plan, and executes the necessary actions to complete your task—no additional input required.

Configuration

Opsy is configured via a YAML file located at ~/.opsy/config.yaml:

# UI configuration
ui:
  # Theme for the UI (default: "default")
  theme: default

# Logging configuration
logging:
  # Path to the log file (default: "~/.opsy/log.log")
  path: ~/.opsy/log.log
  # Logging level: debug, info, warn, error (default: "info")
  level: info

# Anthropic API configuration
anthropic:
  # Your Anthropic API key (required)
  api_key: your_api_key_here
  # Model to use (default: "claude-3-7-sonnet-latest")
  model: claude-3-7-sonnet-latest
  # Temperature for generation (default: 0.5)
  temperature: 0.5
  # Maximum tokens to generate (default: 1024)
  max_tokens: 1024

# Tools configuration
tools:
  # Maximum duration in seconds for a tool to execute (default: 120)
  timeout: 120
  # Exec tool configuration
  exec:
    # Timeout for exec tool (0 means use global timeout) (default: 0)
    timeout: 0
    # Shell to use for execution (default: "/bin/bash")
    shell: /bin/bash

You can also set configuration using environment variables with the prefix OPSY_ followed by the configuration path in uppercase with underscores:

# Set the logging level
export OPSY_LOGGING_LEVEL=debug

# Set the tools timeout
export OPSY_TOOLS_TIMEOUT=180

The Anthropic API key can also be set via ANTHROPIC_API_KEY (without the OPSY_ prefix).

Extending & Contributing

We welcome contributions to Opsy! The project is designed to be easily extended.

To contribute:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Please update tests as appropriate and follow the existing coding style.

Here's how you can extend Opsy's capabilities:

System Prompts

System prompts in ./assets/prompts define how Opsy understands and responds to user tasks:

Agent System Prompt

The primary prompt (assets/prompts/agent_system.tmpl) guides Opsy's overall behavior, establishing its identity as an AI assistant for SREs and DevOps professionals and defining the format for execution plans.

Tool System Prompt

This prompt (assets/prompts/tool_system.tmpl) defines how Opsy interacts with external tools, ensuring interactions are safe, effective, and follow best practices.

Tool User Prompt

This prompt (assets/prompts/tool_user.tmpl) defines the format for requesting tool execution, maintaining consistency in how tools are invoked.

To contribute a new prompt or modify an existing one, add it to the repository and submit a pull request.

Tools

Tool definitions in assets/tools/ allow Opsy to interact with various systems and services:

---
display_name: Tool Name
executable: command-name
description: Description of what the tool does
inputs:
  parameter1:
    type: string
    description: Description of the first parameter
    default: "default-value"  # Optional default value
    examples:
      - "example1"
      - "example2"
    optional: false  # Whether this parameter is required
rules:
  - 'Rule 1 for using this tool'
  - 'Rule 2 for using this tool'

Themes

Theme definitions in assets/themes/ control Opsy's visual appearance:

base:
  base00: "#1A1B26"  # Primary background
  base01: "#24283B"  # Secondary background
  base02: "#292E42"  # Borders and dividers
  base03: "#565F89"  # Muted text
  base04: "#A9B1D6"  # Primary text

accent:
  accent0: "#FF9E64"  # Command text
  accent1: "#9ECE6A"  # Agent messages
  accent2: "#7AA2F7"  # Tool output

Acknowledgments

• Charm for their TUI libraries
• Anthropic for their Go SDK for Claude AI models
• Viper for configuration management
• Various Go libraries for schema validation, data structures, and YAML parsing
• The Go community for excellent tooling and support

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.