Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Rebrand to Anvil AI. Remove CacheForge marketing copy. Normalize install commands.
What This Skill Does
Runs terraform plan and classifies every infrastructure change as safe, moderate, dangerous, or critical. Detects destroys, IAM mutations, data-loss risks, and blast radius before you apply. Never modifies state or runs apply.
Classifies every change by risk level and surfaces destroys and IAM mutations automatically, instead of requiring engineers to manually scan raw plan output.
When to use it
- Checking whether a plan destroys a production database before applying
- Reviewing IAM policy changes in a staging deployment
- Assessing blast radius when replacing a load balancer or VPC subnet
- Validating Terraform config syntax before a team code review
- Identifying which resources will be permanently deleted in a refactor
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Is this plan safe to apply? (directory: ~/infra/prod)
- 1Runs tf-plan-review.sh plan against ~/infra/prod
- 2Parses JSON output for overall_risk, destroy count, and per-resource risk classifications
- 3Identifies critical changes such as IAM role deletions, database replacements, and security group updates
- 4Assesses blast radius for destroyed or replaced resources
- 5Presents Markdown risk report with a go/no-go recommendation
Risk report showing overall risk level, a list of dangerous changes with plain-English explanations, and a pre-apply checklist
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.