- Home
- Skills
- DevOps & Cloud
- Terraform Replacement & Destruction Review Card
Terraform Replacement & Destruction Review Card
Turn exported Terraform plan JSON into a local review queue for planned deletes and replacements.
$7
Works with the AI Agents you already use
Terraform Replacement & Destruction Review Card
Example session with this skill installed
Use the included invented examples/plan.json. Run the local review into a fresh output directory. Show the destructive-action queue and tell me which items require human review, without judging whether the plan is safe to apply.
- Read your context and instructions
- Compiled the terraform replacement
Terraform plan destructive-action review
Status: REVIEW. The included sample plan has 3 resource changes: 1 DESTROY, 1 REPLACE_DESTROY_FIRST, and 1 UPDATE. These are observations from the supplied JSON, not a prediction of the live infrastructure state.
| Address | Planned actions | Classification | Supplied action reason |
|---|---|---|---|
| example_storage.cache | delete | DESTROY | delete_because_no_resource_config |
| example_instance.old | delete, create | REPLACE_DESTROY_FIRST | replace_because_cannot_update |
The UPDATE is counted but is not classified as safe. The packet also includes terraform-change-card.json, HUMAN_REVIEW.md, and evidence-receipt.json. The review card omits resource before/after values; the original plan may still contain secrets and must remain private.
Next steps for the human reviewer: confirm that this is the intended, current plan artifact; inspect the full Terraform plan for each listed address, action ordering and reason; separately evaluate create/update and provider-specific effects. Do not apply infrastructure based on this card alone. A stale plan or omitted context can change the real outcome.
Connects securely to your tools. The creator never sees your data.
About this skill
Export a saved Terraform plan with terraform show -json, then run the local CLI or optional stdio MCP tool on that JSON file. It separates planned deletion, destroy-then-create and create-before-destroy replacement, counts other actions, and writes a deterministic review card and integrity receipt. The companion Skill guides the human reviewer. It never invokes Terraform or a cloud API and does not copy before/after resource values into the card.
Includes installable Python source, CLI, optional local MCP adapter, Skill, invented plan JSON and sample packet. Supports Python 3.11–3.14; optional MCP mode installs the pinned official SDK.
This is an offline queue for actions observed in one supplied plan export. It does not prove the plan is current, safe or complete. Create/update actions can also have consequences. Keep the original plan private because it may contain secrets. Read one selected local JSON file and create one new local output directory; no Terraform execution, cloud credentials, network or infrastructure changes.
Only resource_changes in the supplied Terraform plan JSON are inspected. A plan can become stale; this card does not apply, refresh, or validate live infrastructure. Create/update/read actions are counted but not judged safe. The original plan may contain secrets.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Requires
Permissions
Reads a user-selected local Terraform plan JSON and writes a new local output directory. Shell runs the bundled CLI. No network, cloud credentials, or Terraform apply.