Dagger (16,132 Stars) CI/CD Engine 2026: Write Pipelines in Python or Go, Run Anywhere

Dagger (16,132 stars) lets you write CI/CD pipelines as code in Python, Go or TypeScript - the same pipeline runs locally, in GitHub Actions, or in any runner. Here is the setup.

💡 What You Will Learn

Dagger (16,132 stars) lets you write CI/CD pipelines as code in Python, Go or TypeScript - the same pipeline runs locally, in GitHub Actions, or in any runner. Here is the setup.

## The short answer **dagger/dagger** (16,132 stars, Go) is an automation engine for building, testing, and shipping any codebase. Instead of writing YAML that only works inside one CI vendor, you write pipelines in a real programming language. The pipeline runs the same way locally, in CI, or on any machine with a container runtime. ## Why it is different - **Real languages**: Python, Go, TypeScript - full IDE support, tests, and reusable functions - **Same pipeline everywhere**: debug locally, then run unchanged in GitHub Actions, GitLab CI, or Jenkins - **Container-native**: every step runs in a hermetic container, so builds are reproducible - **Caching built-in**: content-addressed caching makes re-runs fast ## Minimal Python pipeline ```python import dagger async def main(): async with dagger.Connection() as client: src = client.host().directory(".") python = ( client.container() .from_("python:3.12") .with_directory("/src", src) .with_workdir("/src") .with_exec(["pip", "install", "-r", "requirements.txt"]) .with_exec(["pytest"]) ) await python.stdout() import asyncio asyncio.run(main()) ``` Install the SDK and run it: ```bash pip install dagger-io python pipeline.py ``` ## Real-world usage tips - Start by containerizing your build steps - Dagger shines when every stage is a container. - Use `with_exec` for steps and `cache_volume` for package caches (pip, npm). - Keep pipelines in a `ci/` folder and version them with your code. ## FAQ **Does it replace GitHub Actions?** Not entirely - you can use Dagger inside GitHub Actions as the pipeline engine, keeping triggers and hosting there. **Do I need Docker?** A container runtime is required (Docker, containerd, or a remote engine). **Is it production-ready?** Yes - it is used in production by many companies and the engine is Apache-2.0.