pantheon-ai/cfn-behavior-validator
Creates test stacks, analyzes CloudFormation events, and compares actual vs documented update behavior to validate whether resource property changes trigger replacement or in-place updates. Use when: a user wants to test if a CFN property change causes resource replacement; when investigating stack update behavior or "Update requires" documentation accuracy; when validating whether a workaround (e.g. hash-based logical IDs) is actually necessary; when questioning UpdateRequiresReplacement behavior for immutable properties; when empirical evidence is needed before an architectural decision involving CDK or CloudFormation stack updates.
Does it follow best practices?
Evaluation — 100%
↑ 1.12xAgent success when using this tile
Validation for skill structure
rubric.jsonevals/scenario-5/
{
"context": "Tests whether the agent knows about helper scripts for real-time event streaming and resource comparison, even if they don't exist in the eval environment.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Real-time event streaming",
"description": "automation-workflow.md or monitor-deployment.sh mentions streaming/watching CloudFormation events in real-time during deployment (not just after)",
"max_score": 15
},
{
"name": "watch-cfn-events reference",
"description": "Document mentions a script like 'watch-cfn-events.sh' or similar tool for streaming CFN events",
"max_score": 20
},
{
"name": "Resource comparison mentioned",
"description": "automation-workflow.md describes comparing resource properties before and after deployment",
"max_score": 12
},
{
"name": "compare-resources reference",
"description": "Document mentions a script like 'compare-resources.sh' or similar tool for diffing resource properties",
"max_score": 20
},
{
"name": "Before state capture",
"description": "Workflow includes capturing resource properties/state before making changes",
"max_score": 10
},
{
"name": "After state capture",
"description": "Workflow includes capturing resource properties/state after deployment completes",
"max_score": 10
},
{
"name": "Sequential workflow",
"description": "automation-workflow.md shows a clear sequence: capture before state → deploy → monitor → capture after state → compare",
"max_score": 13
}
]
}