Docker

Build and run the Nylas CLI in a container.

Quick Links: README | Development | Security

---

Build Locally

docker build -t nylas-cli:dev .
docker run --rm nylas-cli:dev --version
docker run --rm nylas-cli:dev commands --json

The image runs as the unprivileged nylas user with NYLAS_DISABLE_KEYRING=true (containers don't have a keyring).

---

Usage

Pass your API key as an environment variable and run commands directly:

docker run --rm \
  -e NYLAS_API_KEY="$NYLAS_API_KEY" \
  ghcr.io/nylas/cli:latest \
  email list --limit 5 --json

For commands that need a grant ID:

docker run --rm \
  -e NYLAS_API_KEY="$NYLAS_API_KEY" \
  -e NYLAS_GRANT_ID="$NYLAS_GRANT_ID" \
  ghcr.io/nylas/cli:latest \
  calendar list --json

To target a different API environment (staging, EU):

docker run --rm \
  -e NYLAS_API_KEY="$NYLAS_API_KEY" \
  -e NYLAS_API_BASE_URL="https://api.eu.nylas.com" \
  ghcr.io/nylas/cli:latest \
  email list --limit 5 --json

Do not bake API keys or credentials into the image.

---

Persistent Configuration (Optional)

For repeated use, you can run init with mounted volumes so config survives between runs:

docker volume create nylas-config
docker volume create nylas-cache

docker run --rm -it \
  -v nylas-config:/home/nylas/.config/nylas \
  -v nylas-cache:/home/nylas/.cache/nylas \
  ghcr.io/nylas/cli:latest \
  init

Then reuse the volumes for later commands without passing env vars each time.

---

Web Interfaces

nylas air, nylas ui, and nylas chat bind to localhost inside the container, so published ports won't expose them to the host. Use Docker for CLI commands only.

---

Release Images

Tagged releases publish multi-platform images to GitHub Container Registry:

docker pull ghcr.io/nylas/cli:latest
docker pull ghcr.io/nylas/cli:1.5.0

The release workflow packages GoReleaser Linux artifacts into linux/amd64 and linux/arm64 images.