Qodo Build Your Own Agent Challenge

Developers are no longer just writing code—they’re managing interconnected workflows that stretch across code review, testing, CI/CD, release documentation, and beyond. At Qodo, we built Qodo Command to give developers a flexible way to create and run AI-powered agents across their software development lifecycle.

Now, we’re excited to announce the Build Your Agent Campaign, a challenge designed to spotlight the most innovative agents you can build with Qodo Command.

Why We’re Launching This Campaign

Qodo Command gives developers full control over how agents are created, when they run, and what tools they use. Whether it’s generating tests, reviewing pull requests, fixing issues automatically, or orchestrating multi-repo workflows, these agents are designed to reduce friction, raise code quality, and scale your development processes.

Read more about Qodo Command in this blog.

The Build Your Agent Challenge is your chance to put those capabilities to work, showcase your creativity, and win prizes.

What is Qodo Command?

Qodo Command is our CLI interface for building, running, and automating AI agents anywhere in your software development lifecycle. With Qodo Command, you can:

• Define custom agents you control – Decide what your agent does, when it runs, and the tools it uses.
• Automate workflows across your SDLC – Trigger agents from CI/CD jobs, webhooks, or run them as MCPs.
• Interact from terminal or browser – Use agents directly in your terminal or spin up a web UI for a more visual experience.

Qodo Command uses a .toml configuration format where you define an agent’s:

• Trigger (when it runs, like a PR opened or a CI failure)
• Input (the data it consumes)
• Action (what it does with that data)
• Result (the output or changes it delivers)

This makes it easy to create agents that are reusable, versioned, and seamlessly integrated into your workflows, whether you’re reviewing code, fixing bugs, generating tests, or managing releases. See reference implementations for agents Qodo developers created themselves: https://github.com/qodo-ai/agents 

You can build agents for:

• Code Quality: Agents that run static analysis, enforce style, generate or expand tests, flag vulnerabilities, or remediate issues automatically.
• Quality Automation: Agents that connect CI/CD, GitHub, Slack, or webhooks to automate quality checks across systems and pipelines.
• Performance & Optimization: Agents that audit applications, detect bottlenecks, and suggest improvements.
• Developer Experience: Agents that streamline onboarding, triage errors, or reduce repetitive tasks.

And our community has already started.

One example is the Package Health Reviewer, an open-source agent that automates package health checks using Snyk Advisor. It integrates seamlessly into CI/CD workflows, analyzing security, maintenance, and community metrics of dependencies to produce clear health scores—“healthy,” “sustainable,” or “risky.” This empowers developers to make informed choices about their dependencies and maintain secure, reliable projects. It’s a perfect example of how Qodo agents can solve real-world challenges in modern development.

Other early community-built agents include:

• Accessibility Auditor – checks websites for accessibility issues.
• Performance Optimizer – speeds up sites by diagnosing slowdowns.
• Batch Playwright Fixer – automatically runs and fixes tests.
• Rails Component Generator – scaffolds new components with best practices built-in

How the Build Your Agent competition works

Step 1: Build Your Agent

Install Qodo Command:

npm install -g @qodo/command

• Define your agent using our .toml configuration format.
• Explore our agent repo for examples .

Step 2: Submit Your Agent

1. Open a pull request in the agent repo
2. (Optional) Create a short demo video.
3. Post your agent on X with the hashtag #QodoAgentChallenge and tag @QodoAI.
4. Share your use case via the campaign’s Google form

Prizes and Categories

We’ll be awarding $2,000 in each of four categories,  each spotlighting how developers are using Qodo Command to improve code quality and reliability.

Best Agent for Clean Code

For agents that keep projects maintainable and easy to read over time. This could mean enforcing style guides, reducing duplication, simplifying complex functions, or automatically improving documentation so teams can focus on building rather than deciphering messy code.

Best Agent for Reliable Code

For agents that strengthen correctness and prevent errors. Examples include agents that generate or expand test coverage, validate logic before it ships, monitor for regressions, or automatically fix common issues so developers can ship with confidence.

Best Agent for Quality Automation

For agents that embed quality into the development pipeline itself. These agents integrate checks into CI/CD, orchestrate tools across systems, and automate remediation so that code quality isn’t a one-time task but an ongoing, automated process.

Best Agent for Complex Codebases

For agents that can navigate and automate work across large, multi-service, or enterprise-scale projects. These might help developers coordinate changes across repositories, manage dependencies, or handle large-scale migrations without getting lost in the complexity.