Skip to main content
qckfx (pronounced “quick fix”) is an open-source AI coding platform that offers two distinct products for different use cases:

​
🚀 qckfx CLI - Headless AI Coding Assistant

Perfect for terminal-based workflows and automation The qckfx CLI is designed for developers who want a powerful AI coding assistant that can run headlessly in terminals, CI/CD pipelines, or GitHub Actions. It’s optimized for automation and can operate with minimal human oversight. 
npm install -g @qckfx/agent
qckfx "fix the type errors in the auth module"
Key capabilities:
  • Natural language commands from terminal
  • Headless operation for CI/CD and automation
  • Custom agent configurations via JSON files
  • Git operations, file editing, and code analysis
  • Can run locally (with appropriate caution) or in sandboxed environments
⚠️ Security Note: While the CLI can run in your local terminal, it’s designed for headless automation. When running locally, use appropriate caution as it can make real changes to your codebase.

Get Started with CLI

Install the CLI and run your first AI coding task

​
🛠️ qckfx SDK - Embedded AI Agent Platform

Perfect for building applications with integrated AI coding capabilities The qckfx SDK is a TypeScript/JavaScript library for developers who want to embed AI coding agents directly into their applications, providing full programmatic control over agent behavior. 
import { Agent } from '@qckfx/agent';

const agent = new Agent({
  defaultModel: 'anthropic/claude-sonnet-4',
  systemPrompt: 'You are a helpful AI coding agent...'
});

const result = await agent.processQuery('What does this repository do?');
Key capabilities:
  • Full programmatic control over agent lifecycle
  • Streaming updates and event handling
  • User permission systems and checkpoints
  • Custom tool configurations
  • Integration into web apps, desktop apps, or services

Get Started with SDK

Embed AI coding agents into your applications

​
Which Should You Choose?

Use CLI if you want to:

  • Automate coding tasks in CI/CD pipelines
  • Run AI assistance in GitHub Actions
  • Get quick help from terminal while coding
  • Build headless automation workflows
  • Use pre-built agent configurations

Use SDK if you want to:

  • Build custom applications with AI agents
  • Create user-facing coding assistants
  • Integrate agents into existing software
  • Have full control over agent behavior
  • Build commercial products with embedded AI

​
Shared Foundation

Both products share the same core capabilities:
  • Open source: Full transparency and community-driven development
  • Provider flexibility: Support for any LLM provider with OpenAI-compatible APIs
  • Dynamic codebase understanding: No upfront indexing required
  • Comprehensive toolset: File operations, bash commands, git integration, and more
  • Security focused: Direct API connections and configurable execution environments

​
Quick Tour

Here’s what you can do with qckfx:

​
CLI Examples

# Ask questions about your codebase
qckfx "how does our authentication system work?"

# Fix issues across multiple files
qckfx "fix the type errors in the auth module"

# Automate Git operations
qckfx "commit my changes and create a PR"

​
SDK Examples

// Process queries programmatically
const result = await agent.processQuery('What does this repository do?');

// Handle streaming updates
agent.on('toolCall', (tool) => console.log(`Using ${tool.name}`));

// Control permissions
const agent = new Agent(config, {
  onPermissionRequest: async (request) => await askUser(request)
});

​
Additional Resources

GitHub Actions

Setup qckfx in GitHub Actions for automated AI assistance

Configuration

Customize agents for your specific workflow needs

Troubleshooting

Solutions for common issues with qckfx

Concepts

Understand the core concepts behind qckfx agents

​
License and Data Usage

qckfx is provided under the MIT license.

​
Feedback transcripts

If you choose to send us feedback about qckfx, such as transcripts of your usage, we may use that feedback to debug related issues and improve qckfx’s functionality (e.g., to reduce the risk of similar bugs occurring in the future). We will not train generative models using this feedback.

​
License

© qckfx, Inc 2025. All rights reserved. Open source under the MIT License - free to use, modify, and distribute.
⌘I