Skip to content
← Tools / Skills
ToolDagger/10 trustVerified

Dagger

Programmable CI/CD pipelines in real code — write them once in Go, Python, or TypeScript and run anywhere.

Visit official site →

About

Dagger is a programmable engine for CI/CD pipelines, founded by Docker co-creator Solomon Hykes with Sam Alba and Andrea Luzzardi. Instead of YAML, you define pipelines as functions in Go, Python, or TypeScript; Dagger executes each step in containers, which means a pipeline that passes locally behaves the same in any CI system. It raised $30 million (including a $20M Series A led by Redpoint) and is open source. The Daggerverse hosts reusable community modules, and it integrates with GitHub Actions, GitLab, Jenkins, and others.

The honest picture: Dagger is a pipeline engine, not a deployment orchestrator — teams that expect it to manage SSH deploys or blue-green cutovers end up rebuilding those themselves. Running the Dagger engine alongside Docker adds real overhead on developer machines, its error messages from strict type reflection can be cryptic, and local-first debugging of remote pipelines is still rough. It shines for build/test/release pipelines, less so for the last mile of deployment.

What it can do

  • Pipelines defined in Go, Python, or TypeScript — no YAML
  • Containerized execution: local runs match CI runs
  • Daggerverse: reusable community pipeline modules
  • Strong caching and parallelism via BuildKit
  • Portable across GitHub Actions, GitLab, Jenkins, and more

Setup

Install the Dagger CLI, then write pipeline functions using the Go, Python, or TypeScript SDK. Run 'dagger call' to execute locally, or wire it into your existing CI runner.

Prerequisites

Docker (or a compatible runtime) on the machine running pipelines; familiarity with Go, Python, or TypeScript.

Use cases

ci-cddevopsbuild-automationcontainers

Works with

Trust score — 8/10

/10 trustVerified

Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 52; Trending score 78/100 from researcher assessment.

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.