Four steps to a running workflow.
The graph you draw compiles to an Argo Workflows manifest. It lands in your repo and runs on the cluster you pick.
An Argo Workflows CI template, imported.
1. Draw it, or bring what you already run.
Place the steps and draw the dependencies. Or import a manifest from a repo, a live cluster, or a paste.
2. Read the manifest while you work.
The compiled YAML sits in a pane next to the graph, and updates as you edit. kubectl and argo submit take it as an ordinary manifest.
3. Send it through review.
dagweave writes to the GitHub repo and path you choose, as a direct commit or a pull request. Several files, one commit. Your reviewer never sees a half-updated repo.
4. Dispatch it to whichever cluster you choose.
An operator enrolls each cluster once, with one helm install. dagweave sends the run to the cluster you name at dispatch time. Phase, per-step status, logs and signed artifact URLs come back the same way. Your own Argo Workflows controller runs it.
Change one field, get a one-line diff.
Import a manifest, change one field on the canvas, publish. What your reviewer sees is the field you changed. Everything you didn't touch comes back out exactly as it went in.
kind: Workflow
metadata:
generateName: hello-
spec:
entrypoint: whalesay
templates:
- name: whalesay
container:
image: docker/whalesaykind: Workflow
metadata:
generateName: hello-
spec:
entrypoint: whalesay
templates:
- name: whalesay
container:
image: docker/whalesay
command: [cowsay] # the one thing you changedAn imported manifest opens whole. The retry strategy, the artifact repository ref and the withParam loop come back precisely as they went in.
Your whole team can write Argo Workflows.
dagweave is in early access. Leave an email and we will bring you on as we open it up.