Monitor cloud job execution (K8s, Lambda, Cloud Run); on failure, fetch logs, LLM diagnoses root cause (timeout, permission, resource) — alert instantly.
What it does: Webhook (K8s job/Lambda failure event) → extract job_id/run_id/status → IF failed → fetch job logs from cloud provider API → LLM analyzes error type, suggests remediation → Slack alert with diagnosis and logs link.
Connect cloud orchestrator: K8s (API), Lambda (CloudWatch Logs), Cloud Run (Cloud Logging)
Add cloud credentials: AWS/GCP/Azure service account or role
Configure Slack channel: set target channel (default: #cloud-alerts)
Test: trigger job failure; verify diagnosis in Slack
Apps/nodes: Webhook, cloud provider API, Slack.
Credentials required: Cloud orchestrator account credentials, Slack workspace.
Difficulty: Medium | Setup time: 8 minutes.
Business outcome: Engineers see job failure diagnosis instantly, MTTR ↓ 70%, incidents resolved faster.
With an active subscription, download the workflow file right from this page or your dashboard.
Choose Import from File and select the downloaded JSON. The full agent graph appears, ready to configure.
Each integration node prompts for credentials on first run. The setup guide lists every credential the agent expects.
Run once with sample input, confirm the expected output, then flip the activate toggle.
No reviews yet. Be the first.
Sign in to leave a review.
This agent — and every other in the collection — comes with your subscription. One plan, the whole catalog.
Subscribe to downloadOn cloud incident: auto-generate postmortem structure (timeline, root cause, lessons learned, action items); save to doc store and notify team.