Contribution Overview
Contributing to Opik
We’re excited that you’re interested in contributing to Opik! There are many ways to contribute, from writing code to improving the documentation, or even helping us with developer tooling.
How You Can Help
Help us improve by submitting bug reports and suggesting new features.
Improve our guides, tutorials, and reference materials.
Set up your local Opik development environment with Docker, local processes, or manual setup.
Enhance our Python library for Opik.
Help improve our TypeScript library for Opik.
Work on our agent optimization tools.
Develop new features and UI improvements for the Opik web application.
Work on the core Java services powering Opik.
The bounty program is currently paused. Check status updates here.
Speak or write about Opik and share your experiences. Let us know on Comet Chat!
Also, consider reviewing our Contributor License Agreement (CLA).
Submitting a new issue or feature request
This is a vital way to help us improve Opik!
Submitting a New Bug Report
Before submitting a new issue, please check the existing issues to avoid duplicates.
To help us understand the issue you’re experiencing, please provide:
- Clear steps to reproduce the issue.
- A minimal code snippet that reproduces the issue, if applicable. This helps us diagnose the issue and fix it more quickly.
Submitting a New Feature Request
Feature requests are welcome! To help us understand the feature you’d like to see, please provide:
- A short description of the motivation behind this request.
- A detailed description of the feature you’d like to see, including any code snippets if applicable.
If you are in a position to submit a PR for the feature, feel free to open a PR!
Project Setup and Architecture
Opik is a monorepo with multiple services and SDKs. Common contributor entry points are:
apps/opik-backend: Core Java backend API/servicesapps/opik-frontend: React frontend applicationapps/opik-documentation: Documentation website and docs sourceapps/opik-guardrails-backend,apps/opik-python-backend,apps/opik-sandbox-executor-python: supporting backends and runtime servicessdks/python,sdks/typescript,sdks/opik_optimizer: SDKs and optimizer toolingtests_end_to_end: E2E suites and helper services
Opik relies on: ClickHouse (traces, spans, feedback), MySQL (metadata), and Redis (caching).
The local development environment uses convenient scripts (./opik.sh for Linux/Mac, .\opik.ps1 for Windows) that manage Docker Compose automatically. Please see instructions in the deployment/docker-compose/README.md on GitHub for advanced usage.
Developer Tooling & AI Assistance
To help AI assistants (like Cursor) better understand our codebase, we provide context files:
- General Context:
https://www.comet.com/docs/opik/llms.txt- Provides a general overview suitable for most queries. - Full Context:
https://www.comet.com/docs/opik/llms-full.txt- Offers a more comprehensive context for in-depth assistance.
You can point your AI tools to these URLs to provide them with relevant information about Opik.
AI-Assisted Contributions (Required Disclosure)
AI assistance and authorship is allowed. Human authors remain fully accountable for correctness, licensing, and security.
Rules:
- Always run relevant tests/linters for touched code.
- Always be explicit about human/users interaction with produced output.
- Always review prior issue, pull-requests and code-base for existing solutions.
- Always address any system generated reviews (Baz, Greptile).
- Never submit unreviewed AI output.
- Never include secrets, tokens, private prompts, internal system instructions, or customer-sensitive data in generated/public content.
- Never disclose vulnerabilities, exploit steps, or incident details in public issues/PRs. Use private maintainer/security channels.
- You must ensure every PR include an AI disclosure watermark block in the PR description: If AI is used, include a brief
## AI Assistancenote in the PR description with tool/model, scope, and confirmed human verification (if any); if no AI is used, omit that section entirely.
Review our Contributor License Agreement (CLA) if you haven’t already. Comment on popular feature requests to show your support.