terraform-cloud-mcp

# Plan Management Conversation Examples This document provides example conversations demonstrating how to use the plan management tools in the Terraform Cloud MCP. ## Getting Plan Details ``` **User:** I'd like to see details for my plan with ID plan-aBcdEfGh12345678 **Claude:** I'll retrieve the details for that plan for you. [Claude uses get_plan_details tool] Here are the details for plan-aBcdEfGh12345678: Status: finished Resources to add: 3 Resources to change: 1 Resources to destroy: 0 The plan was queued at 2025-02-15T14:30:22Z, started at 2025-02-15T14:31:05Z, and finished at 2025-02-15T14:32:18Z. It's associated with run-pQrStUvW12345678 in workspace "production-infrastructure". ``` ## Retrieving Plan Logs ``` **User:** Can you show me the actual Terraform plan output for plan-aBcdEfGh12345678? **Claude:** I'll retrieve the plan logs for you. [Claude uses get_plan_logs tool] I've retrieved the complete plan execution logs. The plan shows that Terraform will create an S3 bucket called "example-data-bucket" with versioning enabled and production tags, create a new IAM role for Lambda execution, and modify an existing security group to add an HTTPS ingress rule allowing traffic from anywhere. The plan execution used Terraform v1.5.7 and successfully determined that 2 resources will be added and 1 will be changed, with no resources being destroyed. Would you like me to explain any part of this plan in more detail? ``` ## Retrieving Plan JSON Output ``` **User:** Can you get me the JSON execution output for plan-aBcdEfGh12345678? I need to analyze the changes. **Claude:** I'll retrieve the JSON execution plan for you. [Claude uses get_plan_json_output tool] I've obtained the JSON execution plan. Here's an analysis of the planned changes: The plan includes: - 2 resources to be created (aws_s3_bucket.data and aws_iam_role.lambda_role) - 1 resource to be modified (aws_security_group.allow_http) - 0 resources to be destroyed Key modifications: - The security group is having a new ingress rule added for HTTPS (port 443) - The new S3 bucket will have versioning enabled - The IAM role has permissions for Lambda and S3 access The plan was generated with Terraform 1.5.7 and includes all configuration details and state information. Would you like me to extract any specific part of the plan for more detailed analysis? ``` ## Retrieving Run Plan JSON Output ``` **User:** I'm working with a run ID run-xYzAbCdE12345678 and need the JSON execution plan from that run. **Claude:** I'll get the JSON plan output from that run for you. [Claude uses get_run_plan_json_output tool] I've retrieved the JSON execution plan for run-xYzAbCdE12345678. This is a destruction plan that will remove the following resources: - aws_instance.web_server (EC2 instance) - aws_security_group.web_access - aws_s3_bucket.logs - aws_cloudwatch_metric_alarm.high_cpu The plan was generated with Terraform version 1.6.2. All four resources will be completely removed, and the state will be updated accordingly. The plan shows that this will remove: - 1 EC2 t2.micro instance in us-west-2 - 1 security group with 3 ingress rules - 1 S3 bucket containing approximately 250MB of log data - 1 CloudWatch alarm that was monitoring the EC2 instance Would you like me to extract any specific details from the plan? ``` ## Working with Plan Status ``` **User:** My plan seems to be taking a long time. The ID is plan-mNoPqRsT12345678. What's its status? **Claude:** Let me check the status of that plan for you. [Claude uses get_plan_details tool] Your plan with ID plan-mNoPqRsT12345678 is currently in the "running" status. It was queued at 2025-03-01T09:15:33Z and started execution at 2025-03-01T09:16:10Z (about 25 minutes ago). This plan is running in an agent execution mode on agent "prod-deployment-agent-05" in the "Production" agent pool. Complex plans with many resources can take time to complete, especially if there are many computed values or external data sources. Would you like me to check if there are any logs available that might provide more information about the progress? ``` ## Analyzing Plan Resources ``` **User:** I need to understand what resources will be created in plan-gHiJkLmN12345678 **Claude:** I'll retrieve and analyze the plan details for you. [Claude uses get_plan_details tool] The plan plan-gHiJkLmN12345678 will create 5 new resources: - 2 aws_security_group resources - 1 aws_instance (EC2) resource - 1 aws_s3_bucket resource - 1 aws_iam_role resource The plan will also modify 2 existing resources and destroy 1 resource. This plan was created as part of run-eFgHiJkL12345678 in workspace "network-infrastructure" and has already finished execution. It's ready to be applied. For more detailed analysis of the exact resource configurations, I can retrieve the JSON execution plan which contains the full definition of all resource changes. ``` ## Comparing Multiple Plans ``` **User:** I have two plans I want to compare: plan-qRsTuVwX12345678 and plan-yZaBcDeF12345678 **Claude:** I'll retrieve both plans and compare them for you. [Claude uses get_plan_details tool] Here's a comparison of the two plans: Plan 1 (plan-qRsTuVwX12345678): - Status: finished - Resources to add: 4 - Resources to change: 2 - Resources to destroy: 1 - Execution time: 1m 45s (2025-03-10) - Run: run-aBcDeFgH12345678 Plan 2 (plan-yZaBcDeF12345678): - Status: finished - Resources to add: 4 - Resources to change: 3 - Resources to destroy: 0 - Execution time: 2m 12s (2025-03-15) - Run: run-iJkLmNoP12345678 The second plan (from March 15) modifies one additional resource compared to the first plan (from March 10), but doesn't destroy any resources. Both plans add the same number of new resources. To see exactly what changed between these plans, I can retrieve and compare the JSON output for both. ``` ## Examining Plan Logs for Errors ``` **User:** I'm concerned about potential issues in plan-kLmNoPqR12345678. Can you check the logs for any warnings or errors? **Claude:** I'll examine the plan logs for warnings or errors. [Claude uses get_plan_logs tool] I've analyzed the plan logs and found the following issues: Warnings: 1. Deprecated attribute: The plan contains a warning about using the deprecated "public_ip" attribute on aws_instance.web. The recommended replacement is "public_ip_address". 2. Provider version constraint: There's a warning that the azurerm provider being used (3.0.1) is older than the recommended minimum version (3.2.0). Potential errors: 1. Data source dependency: The plan shows that aws_vpc.main will be created before aws_subnet.primary, but there's a reference to the VPC ID in the subnet. This could cause issues if the VPC creation fails. 2. Resource validation warning: An S3 bucket name "my-bucket" might not be globally unique, which could cause the apply to fail. The plan itself completed successfully but these issues might cause problems during apply. Would you like me to give recommendations on how to address these concerns? ```