Stored in your browser only. Never sent to our servers. Get a free key
Summary
Paste your terraform plan output and get an instant AI summary: what changes, what risks exist, and what it costs. Free, browser-only, BYO Anthropic API key.
Stored in your browser only. Never sent to our servers. Get a free key
Use terraform plan -no-color to strip color codes. Copy the full output from your terminal.
Plain Summary for a quick overview, Resource Analysis for a full inventory, Cost Impact for billing estimates, Security Review for risk flags.
Copy to a Slack message, PR comment, Confluence page, or Jira ticket. Share a permalink so teammates can see the same output.
Paste the plan into your PR comment so reviewers see what changes at a glance — no local apply needed.
Present plain-English infrastructure changes to non-Terraform stakeholders without walking them through HCL.
Before a risky apply, generate a Security Review to catch open ingress rules or destroyed persistent resources.
Generate Cost Impact Notes and attach them to Jira tickets for finance team approval before provisioning.
Run terraform plan -no-color to strip ANSI escape codes before pasting. If you're in a CI pipeline, set TF_CLI_ARGS=-no-color in your environment variables. For JSON output, run terraform show -json plan.tfplan after saving the plan file — the AI can read either format.
Plan Summary gives a plain-English paragraph overview suitable for a PR comment or Slack message. Resource Change Analysis lists every resource with add/change/destroy labels and explains what changes within each. Cost Impact Notes estimates billing implications of instance type changes, scaling, or new services. Security Review flags risky changes: open ingress rules, IAM permission grants, public S3 buckets, and destructive operations on persistent resources.
Yes. Terraform marks forced replacements with -/+ (destroy then create). The AI specifically looks for these and flags them — particularly dangerous for stateful resources like RDS instances, ElastiCache clusters, or Kubernetes PersistentVolumeClaims, where recreation means data loss unless you have a snapshot strategy.
Yes. Copy the plan output from the CI log — Atlantis, GitHub Actions, Spacelift, and Terraform Cloud all output standard terraform plan format. Strip any surrounding workflow log lines and paste just the plan section. The AI handles partial or truncated plans too, though a complete plan gives better results.
Yes. OpenTofu (the Terraform fork) produces identical plan output format. Paste it the same way — the AI doesn't distinguish between them.
The plan output goes directly from your browser to Anthropic's API using your own key. Nothing is stored on our servers — there is no backend. That said, terraform plan output contains your resource names, ARNs, and configuration values. If your organization has strict data classification policies, check whether sending plan output to an LLM is compliant, or use a redacted plan.
A typical plan (10-30 resources) costs roughly $0.002–$0.008 with Claude Haiku, or $0.01–$0.03 with Claude Sonnet. You pay Anthropic directly — no subscription, no markup from us.