Full run history
Every invocation logs against the same job history as your dashboard.
Trigger jobs, rerun a single failing test, and get results back in the channel — right from the release conversation your team is already having.
Shipping the auth fix — can we knock the integration suite out before standup?
How it works
Configure Slack integration in the DevAssure web portal.
Type /devassure run-job or the short alias /da and pick a job, or pass an id directly.
Pass, fail, and pending counts post back to the thread as the run completes.
Product demo
Built for slack channels
Every command works in a DM or any channel in your workspace.
Every invocation logs against the same job history as your dashboard.
A failing run posts back with enough detail to act on immediately.
Keep your pipeline triggers as-is; Slack becomes an additional way in.
No local agent, browser, or CLI needed to trigger a run from Slack.
/devassure run-jobpick a job and run it/devassure run-job <id>run a job by id/devassure run-test <id>rerun a single test/dashort alias, same commandsPass, fail, and pending counts land in the same thread the run was triggered from — visible to everyone already in the conversation.
Before and after
The run, its status, and the discussion about it live in the same thread.
Two-minute setup. No infrastructure to own. Just the commands your team already runs, from the channel they're already in.
FAQ
What teams ask before adding the app to a workspace.
No. Connect Slack Workspace to DevAssure, authorize the workspace, and the commands are live — no local agent, browser, or CLI required to trigger a run.
All channels in your workspace. Scope it to a single #release-checks channel or open it up workspace-wide.
Yes. Every /devassure run is logged against the same job history as runs triggered from the dashboard, your CI pipeline, or the CLI — one history, three ways in.
Yes — /devassure run-test <test-id> reruns a single test and posts the result back to the same thread, so you're not waiting on the full suite to chase one failure.
The agent posts back with what failed and a link to the full report — enough to act on without leaving the channel to find out what broke.