Open Source Contributions

How to contribute to Livepeer’s open source repositories — from documentation fixes to protocol engineering — including a guide to every core repo and its contributing process.

Livepeer is fully open source. Every core component — from the protocol node to the AI runner to the documentation — is publicly maintained on GitHub and welcomes contributions from anyone.

You do not need to write code to make a valuable contribution. Documentation improvements, bug reports, test coverage, and community support are all valued.

github.com/livepeer

The official Livepeer GitHub organisation hosting all core repositories.

Ways to Contribute

Contribution Principles

These apply across all Livepeer repositories:

Core Repositories

go-livepeer -- Official Go Protocol Node

The reference implementation of the Livepeer protocol in Go. This is the core node software used by both Orchestrators and Broadcasters (Gateways).

• Language: Go
• Difficulty: Intermediate to advanced. The codebase spans multiple packages. New contributors are encouraged to start with documentation issues or small, well-scoped bug fixes before tackling multi-package changes.
• Contributing guide: github.com/livepeer/go-livepeer/blob/master/CONTRIBUTING.md
• Good first issues: Labelled help wanted and good first issues on the Issues page.
• Changelog rule: Every PR must include an update to CHANGELOG_PENDING.md in the format \#xxx @contributor.

https://github.com/livepeer/go-livepeer

Livepeer Studio -- Dashboard and API

The web-based studio and API for broadcasting, video management, and Livepeer network access. A TypeScript monorepo.

• Language: TypeScript (monorepo with multiple packages)
• Shell requirement: A Unix shell is required (Windows users should use WSL).
• Dependency manager: Use Yarn — not npm.
• Difficulty: Intermediate. The monorepo structure adds cognitive load. First contributions are best directed at a single subpackage.
• Contributing guide: github.com/livepeer/studio/blob/master/CONTRIBUTING.md
• Discussions: Open a Discussion in the repo if you’re unsure about an approach before opening a PR.

https://github.com/livepeer/studio

ai-runner -- AI Inference Runtime

The AI runner processes inference jobs on the Livepeer network. It is a containerised Python application that loads models into GPU memory and exposes a REST API to the Livepeer node AI worker.

• Language: Python (with Go bindings generated from OpenAPI spec)
• Key concepts: Docker containerisation, GPU inference, pipeline architecture
• Contributing guide: See README.md and the development documentation in the repo.
• Note: The AI network is in its Beta phase. Report issues to the Livepeer Discord.

https://github.com/livepeer/ai-runner

ComfyStream -- Real-Time AI Video for ComfyUI

An open-source ComfyUI custom node for running real-time media workflows, enabling AI-powered video and audio processing using ComfyUI as a backend inference engine.

• Language: Python
• Documentation: docs.comfystream.org
• Key concepts: ComfyUI node system, DAG-based pipelines, WebRTC streams, real-time video processing
• Contributing guide: See README.md in the repository.
• Deployment options: Docker image, RunPod template, Tensordock via Python script, Ansible playbook.

https://github.com/livepeer/comfystream

lpms -- Livepeer Media Server

The Livepeer media server library responsible for video transcoding operations.

• Language: Go (with Nvidia GPU integration)
• Contributing guide: See README.md in the repository.
• Note: GPU-specific testing requires Nvidia hardware and tags (--tags=nvidia).

https://github.com/livepeer/lpms

ui-kit -- Video Primitives for React

A React component library providing video UI primitives for building Livepeer-powered video applications.

• Language: TypeScript / React
• Contributing guide: Contributing docs — read before submitting a PR.

https://github.com/livepeer/ui-kit

livepeer-ai-js -- JavaScript/TypeScript AI SDK

The JavaScript and TypeScript client library for the Livepeer AI API, providing type-safe access to generative AI pipelines.

• Language: TypeScript
• Use case: Building AI-powered video applications with Livepeer’s inference infrastructure.

https://github.com/livepeer/livepeer-ai-js

livepeer-ai-python -- Python AI SDK

The Python client library for the Livepeer AI API.

• Language: Python
• Use case: Integrating Livepeer AI inference into Python applications and workflows.

https://github.com/livepeer/livepeer-ai-python

docs -- This Documentation

The Livepeer documentation repository, built on Mintlify with MDX content.

• Language: MDX, JSX, TypeScript (tooling/scripts)
• Contributing guide: See Contribute to the Docs for the full workflow including pull request steps, review timeline, and style guidance.
• Good for: Non-code contributors comfortable with Markdown or Git.

https://github.com/livepeer/docs

First Contribution Workflow

1. Choose a repository

   Start with a repository in a language or area you’re comfortable with. If you’re new to Livepeer, the docs repo or the ui-kit are lower-friction starting points than go-livepeer or ai-runner.

2. Read the CONTRIBUTING.md

   Every Livepeer repository has a CONTRIBUTING.md file. Read it before opening issues or PRs. It covers code style, PR conventions, changelog requirements, and testing expectations.

3. Find an issue

   Browse the Issues tab and filter by good first issue or help wanted. These labels mean the maintainers have identified the issue as suitable for a new contributor and are willing to guide you through it.

4. Comment before you start

   Leave a comment on the issue saying you’re planning to work on it. This prevents duplicate work and opens a conversation with maintainers if the scope is unclear.

5. Fork, branch, and work

   Fork the repository, create a dedicated branch for your change, and make your commits. Keep commits focused and messages descriptive.

6. Open a pull request

   Open a PR against the correct base branch (usually master or main). Fill out the PR template fully. Link the issue your PR addresses. Include changelog entries where required.

7. Engage with review

   Respond to reviewer comments promptly. Push new commits to address feedback instead of force-pushing. Mark conversations as resolved once addressed.

If you still have questions after reading a repo’s contributing guide, open an issue or ask in the Livepeer Discord. The team is generally responsive to genuine first-time contributor questions.

Code of Conduct

All Livepeer repositories follow a community code of conduct based on the Contributor Covenant. Contributors are expected to be constructive, kind, and patient. Review the code of conduct linked in each repo before participating.

Bug Bounties

For security researchers: report smart contract vulnerabilities and earn USDC rewards via Immunefi.

Grants & Programmes

Apply for a grant to fund your open source work on the Livepeer ecosystem.

Contribute to the Docs

Full PR workflow, review process, and style guide for the documentation repository.

Community Guidelines

The values and conduct standards that apply across all Livepeer community spaces.