BC terraform-plan-reviewer
Review terraform plan output for risk before apply. Detects destructive changes (force replace and destroy on stateful resources), drift from real state, missing imports, secrets leaked into outputs or state, IAM permission widening, public-internet exposure on storage and security groups, and provider-specific footguns across AWS, GCP, and Azure. Produces a P0/P1/P2 severity matrix, applies gating rules (block, require-approval, allow), and writes a PR-comment-style review with line citations and suggested fixes. Use when asked to review a terraform plan, gate a CI apply, audit a Terragrunt run, check infrastructure changes for risk, or set up a plan-review GitHub Action. Triggers on "terraform plan", "tf plan", "terragrunt", "tofu plan", "infrastructure as code", "iac review", "plan output", "destroy", "force replace", "iam widening", "drift", "tfstate", "atlantis", "spacelift", "env0".
Review terraform plan output for risk before apply.
As a process C 63/100 · Has gaps — weak spots: inputs and preconditions, failures and branches, running it twice
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5430 tokens (recommended < 5000); move details to references/ - note
edit-residuethe text marks something as outdated (lines 143, 171): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 63/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 13 mutating operations with no state check
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 5430 tokens
- 100Tools and files. No external tools needed
- 100Steps. 45 steps
- 100Consistency. Name and required fields are in place
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 901: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +5Description quotes 13 example trigger phrases
- +4Structure: 26 headings
- +3Step-by-step instructions: 45 items
- +3Output format is stated explicitly
- +4Has examples (19 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.