# AWS FinOps MCP Server - Tools Reference
Complete reference for all 23 tools available in the AWS FinOps MCP Server.
## Table of Contents
- [Cleanup Tools (9)](#cleanup-tools)
- [Capacity Tools (4)](#capacity-tools)
- [Cost Optimization Tools (5)](#cost-optimization-tools)
- [Application Performance Tools (2)](#application-performance-tools)
- [Upgrade Tools (1)](#upgrade-tools)
---
## Cleanup Tools
### 1. find_unused_lambda_functions
**Purpose**: Identify Lambda functions with no invocations in the specified period.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 90): Lookback period in days
- `max_results` (int, default: 100): Maximum results to return
- Plus authentication parameters
**Returns**:
```json
{
"id": 101,
"name": "Unused Lambda Functions",
"count": 5,
"resource": [
{
"FunctionName": "old-function",
"Runtime": "python3.8",
"MemorySize": 128,
"LastModified": "2023-01-15T10:30:00.000+0000",
"Description": "Lambda function not invoked in the last 90 days"
}
]
}
```
**Use Case**: Identify Lambda functions that can be safely deleted to reduce costs.
---
### 2. find_unused_elastic_ips
**Purpose**: Find Elastic IPs not attached to any instance.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- Plus authentication parameters
**Returns**:
```json
{
"id": 106,
"name": "Unused Elastic IPs",
"count": 3,
"resource": [
{
"PublicIp": "54.123.45.67",
"AllocationId": "eipalloc-12345678",
"Name": "old-eip",
"Domain": "vpc",
"Description": "Unused Elastic IP - not attached to any instance"
}
]
}
```
**Use Case**: Unattached EIPs incur charges. Release them to save costs.
---
### 3. find_unused_amis
**Purpose**: Find AMIs not used by any EC2 instances, ASGs, or Spot Fleet Requests.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 90): Minimum age in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 107,
"name": "Unused AMIs",
"count": 10,
"resource": [
{
"ImageId": "ami-12345678",
"Name": "old-app-image",
"CreationDate": "2023-01-01T00:00:00.000Z",
"Age": "365 days",
"Platform": "Linux/UNIX",
"Description": "AMI not used by any EC2 instances or Auto Scaling Groups"
}
]
}
```
**Use Case**: Old AMIs and their snapshots consume storage. Clean up to reduce costs.
---
### 4. find_unused_load_balancers
**Purpose**: Find load balancers with no traffic in the specified period.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 90): Lookback period in days
- Plus authentication parameters
**Returns**:
```json
{
"id": 103,
"name": "Unused Load Balancers",
"count": 2,
"resource": [
{
"Name": "app/old-alb/abc123",
"Type": "ALB",
"DNSName": "old-alb-123.us-east-1.elb.amazonaws.com",
"Scheme": "internet-facing",
"VpcId": "vpc-12345678",
"Description": "Load Balancer with no traffic in the last 90 days"
}
]
}
```
**Use Case**: Unused load balancers cost ~$16-25/month. Delete to save costs.
---
### 5. find_unused_target_groups
**Purpose**: Find target groups with no registered targets.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 104,
"name": "Unused Target Groups",
"count": 5,
"resource": [
{
"TargetGroupName": "old-tg",
"Protocol": "HTTP",
"Port": 80,
"VpcId": "vpc-12345678",
"LoadBalancerCount": 0,
"Description": "Target group with no registered targets"
}
]
}
```
**Use Case**: Clean up orphaned target groups for better organization.
---
### 6. find_unused_log_groups
**Purpose**: Find CloudWatch Log Groups with no recent log events.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 90): Lookback period in days
- `max_results` (int, default: 50): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 105,
"name": "Unused Log Groups",
"count": 15,
"resource": [
{
"LogGroupName": "/aws/lambda/old-function",
"LastEventTime": "2023-01-01T00:00:00",
"StoredMB": "1024.50",
"RetentionDays": "Never Expire",
"Description": "Log group with no events in the last 90 days"
}
]
}
```
**Use Case**: Log storage costs $0.03/GB/month. Delete old logs to save costs.
---
### 7. find_unused_snapshots
**Purpose**: Find EBS snapshots not associated with any AMI or volume.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 90): Minimum age in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 108,
"name": "Unused Snapshots",
"count": 20,
"resource": [
{
"SnapshotId": "snap-12345678",
"Description": "Manual snapshot",
"StartTime": "2023-01-01T00:00:00",
"Age": "365 days",
"SizeGB": 100,
"State": "completed"
}
]
}
```
**Use Case**: Snapshots cost $0.05/GB/month. Delete unused ones to save costs.
---
### 8. find_unused_security_groups
**Purpose**: Find security groups not attached to any resources.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 114,
"name": "Unused Security Groups",
"count": 10,
"resource": [
{
"SecurityGroupID": "sg-12345678",
"SecurityGroupName": "old-sg",
"Description": "Old security group",
"VpcId": "vpc-12345678",
"InboundRulesCount": 3,
"OutboundRulesCount": 1
}
]
}
```
**Use Case**: Clean up unused security groups for better security posture.
---
## Capacity Tools
### 9. find_underutilized_ec2_instances
**Purpose**: Find EC2 instances with low CPU and memory utilization (≤20%).
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 30): Lookback period in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 109,
"name": "Underutilized EC2 Instances",
"count": 5,
"resource": [
{
"InstanceId": "i-12345678",
"InstanceName": "web-server-1",
"InstanceType": "t3.large",
"VpcId": "vpc-12345678",
"AvailabilityZone": "us-east-1a",
"Vcpu": 2,
"MemoryGiB": "8.0",
"AvgCPUUtilization": 5.2,
"MaxCPUUtilization": 15.0,
"MaxMemoryUtilization": 18.0,
"Description": "Instance is underutilized. Consider downsizing to save costs."
}
]
}
```
**Use Case**: Downsize underutilized instances to save 30-50% on compute costs.
---
### 10. find_overutilized_ec2_instances
**Purpose**: Find EC2 instances with high CPU or memory utilization (≥80%).
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 30): Lookback period in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 111,
"name": "Overutilized EC2 Instances",
"count": 3,
"resource": [
{
"InstanceId": "i-87654321",
"InstanceName": "api-server-1",
"InstanceType": "t3.small",
"VpcId": "vpc-12345678",
"AvailabilityZone": "us-east-1b",
"Vcpu": 2,
"MemoryGiB": "2.0",
"AvgCPUUtilization": 75.5,
"MaxCPUUtilization": 95.0,
"MaxMemoryUtilization": 88.0,
"Description": "Instance is overutilized. Consider upsizing for better performance."
}
]
}
```
**Use Case**: Upsize overutilized instances to prevent performance issues.
---
### 11. find_underutilized_rds_instances
**Purpose**: Find RDS instances with low CPU utilization (≤20%).
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 30): Lookback period in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 110,
"name": "Underutilized RDS Instances",
"count": 2,
"resource": [
{
"InstanceName": "prod-db",
"InstanceType": "db.r5.xlarge",
"Engine": "postgres",
"EngineVersion": "14.7",
"AvailabilityZone": "us-east-1a",
"MultiAZ": true,
"AvgCPUUtilization": 8.5,
"MaxCPUUtilization": 15.0,
"Description": "RDS instance with low CPU utilization (≤20%) in the last 30 days"
}
]
}
```
**Use Case**: Downsize underutilized RDS instances to save on database costs.
---
### 12. find_overutilized_rds_instances
**Purpose**: Find RDS instances with high CPU utilization (≥80%).
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 30): Lookback period in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 112,
"name": "Overutilized RDS Instances",
"count": 1,
"resource": [
{
"InstanceName": "analytics-db",
"InstanceType": "db.t3.medium",
"Engine": "mysql",
"EngineVersion": "8.0.32",
"AvailabilityZone": "us-east-1b",
"MultiAZ": false,
"AvgCPUUtilization": 78.5,
"MaxCPUUtilization": 92.0,
"Description": "RDS instance with high CPU utilization (≥80%) in the last 30 days"
}
]
}
```
**Use Case**: Upsize overutilized RDS instances to prevent database slowdowns.
---
## Cost Optimization Tools
### 13. get_all_cost_optimization_recommendations
**Purpose**: Get all cost optimization recommendations from AWS Cost Optimization Hub (19 resource types).
**Parameters**:
- `region_name` (str, optional): Filter by AWS region
- Plus authentication parameters
**Returns**: List of 19 resource type objects
**Use Case**: Comprehensive cost optimization analysis across all AWS services.
---
### 14. get_cost_optimization_ec2
**Purpose**: Get EC2 instance cost optimization recommendations.
**Parameters**:
- `region_name` (str, optional): Filter by AWS region
- Plus authentication parameters
**Returns**:
```json
{
"id": 201,
"name": "Cost_Ec2Instance",
"count": 10,
"resource": [
{
"RecommendationId": "rec-12345",
"AccountId": "123456789012",
"Region": "us-east-1",
"ResourceId": "i-12345678",
"ResourceArn": "arn:aws:ec2:us-east-1:123456789012:instance/i-12345678",
"CurrentResourceType": "Ec2Instance",
"RecommendedResourceType": "Ec2Instance",
"EstimatedMonthlySavings": 45.50,
"EstimatedSavingsPercentage": 30.0,
"EstimatedMonthlyCost": 151.67,
"CurrencyCode": "USD",
"ImplementationEffort": "Low",
"RestartNeeded": true,
"ActionType": "Rightsize",
"RollbackPossible": true,
"CurrentResourceSummary": "t3.large",
"RecommendedResourceSummary": "t3.medium",
"LastRefreshTimestamp": "2024-01-15T10:30:00Z",
"RecommendationLookbackPeriodInDays": 14,
"Source": "ComputeOptimizer",
"Description": "Cost optimization: Rightsize for Ec2Instance"
}
]
}
```
**Use Case**: AWS-recommended EC2 instance optimizations based on actual usage.
---
### 15. get_cost_optimization_lambda
**Purpose**: Get Lambda function cost optimization recommendations.
**Parameters**: Same as EC2
**Returns**: Similar structure with Lambda-specific recommendations
**Use Case**: Optimize Lambda memory allocation and architecture.
---
### 16. get_cost_optimization_rds
**Purpose**: Get RDS instance cost optimization recommendations.
**Parameters**: Same as EC2
**Returns**: Similar structure with RDS-specific recommendations
**Use Case**: Optimize RDS instance types and storage configurations.
---
### 17. get_cost_optimization_ebs
**Purpose**: Get EBS volume cost optimization recommendations.
**Parameters**: Same as EC2
**Returns**: Similar structure with EBS-specific recommendations
**Use Case**: Optimize EBS volume types (gp2 → gp3, etc.).
---
## Application Performance Tools
### 18. find_target_groups_with_high_error_rate
**Purpose**: Find target groups with high 5XX error rates.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 7): Lookback period in days
- `error_threshold` (float, default: 5.0): Error rate threshold percentage
- Plus authentication parameters
**Returns**:
```json
{
"id": 115,
"name": "Target Group Error Rate",
"count": 2,
"resource": [
{
"TargetGroupName": "api-tg",
"LoadBalancer": "api-alb",
"Protocol": "HTTP",
"Port": 80,
"TotalRequests": 1000000,
"Total5XXErrors": 75000,
"ErrorRate": "7.50%",
"Description": "Target group has 7.50% error rate (threshold: 5.0%)"
}
]
}
```
**Use Case**: Identify application health issues causing high error rates.
---
### 19. find_target_groups_with_high_response_time
**Purpose**: Find target groups with high response times.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 7): Lookback period in days
- `response_time_threshold` (float, default: 1.0): Threshold in seconds
- Plus authentication parameters
**Returns**:
```json
{
"id": 116,
"name": "Target Group Response Time",
"count": 3,
"resource": [
{
"TargetGroupName": "slow-api-tg",
"LoadBalancer": "api-alb",
"Protocol": "HTTP",
"Port": 80,
"AvgResponseTime": "2.500s",
"MaxResponseTime": "5.000s",
"Threshold": "1.0s",
"Description": "Target group has 2.500s avg response time (threshold: 1.0s)"
}
]
}
```
**Use Case**: Identify performance bottlenecks in applications.
---
## Upgrade Tools
### 20. find_asgs_with_old_amis
**Purpose**: Find Auto Scaling Groups using AMIs older than the specified period.
**Parameters**:
- `region_name` (str, default: "us-east-1"): AWS region
- `period` (int, default: 90): Minimum age in days
- `max_results` (int, default: 100): Maximum results
- Plus authentication parameters
**Returns**:
```json
{
"id": 113,
"name": "ASGs with Old AMIs",
"count": 5,
"resource": [
{
"AutoScalingGroupName": "web-asg",
"AMIId": "ami-12345678",
"AMIName": "web-app-v1.0",
"AMICreationDate": "2023-01-01T00:00:00.000Z",
"AMIAge": "365 days",
"Platform": "Linux/UNIX",
"MinSize": 2,
"MaxSize": 10,
"DesiredCapacity": 4,
"Description": "ASG using AMI that is 365 days old (threshold: 90 days)"
}
]
}
```
**Use Case**: Identify ASGs that need AMI updates for security and performance.
---
## Common Parameters
All tools accept these authentication parameters:
- `profile_name` (str, optional): AWS profile name
- `role_arn` (str, optional): IAM role ARN to assume
- `access_key` (str, optional): AWS access key ID
- `secret_access_key` (str, optional): AWS secret access key
- `session_token` (str, optional): AWS session token for temporary credentials
## Response Format
All tools return a standardized response:
```python
{
"id": int, # Unique tool ID (101-219)
"name": str, # Human-readable tool name
"fields": dict, # Field ID to accessor mapping
"headers": list, # Header definitions for display
"count": int, # Number of resources found
"resource": list # List of resource dictionaries
}
```
## Tool IDs
- **101-108**: Cleanup tools
- **109-112**: Capacity tools
- **113**: Upgrade tools
- **114**: Additional cleanup (Security Groups)
- **115-116**: Application performance tools
- **201-219**: Cost optimization tools (by resource type)
