Cloud FinOps Analyst MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AWS_ACCESS_KEY_ID | No | AWS access key ID for authenticating with AWS APIs | |
| AWS_DEFAULT_REGION | No | Default AWS region (e.g., us-east-1) | us-east-1 |
| AWS_SECRET_ACCESS_KEY | No | AWS secret access key for authenticating with AWS APIs |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_azure_cost_summaryA | Fetches Azure cost breakdown by service and resource group for a given time period. |
| get_azure_idle_resourcesA | Identifies idle/underutilized Azure VMs based on average CPU usage over a time period. |
| detect_azure_cost_anomaliesA | Detects daily Azure cost spikes by comparing each day's spend against the period average. |
| get_gcp_cost_summaryA | Fetches GCP cost breakdown by service and project from BigQuery billing export for a given time period. Use this to answer questions about overall GCP spend, top spending services, or cost by project. |
| detect_gcp_cost_anomaliesA | Detects daily GCP cost spikes by comparing each day's spend against the period average, with a per-service breakdown for anomaly days. |
| get_gcp_idle_resourcesA | Identifies idle GCP Compute Engine VMs (low CPU) and unattached Persistent Disks that are wasting money. |
| get_cost_summaryA | Fetches AWS cloud cost breakdown by service and region for a given time period. Use this to answer questions about overall spend, top spending services, or cost by region. |
| detect_cost_anomaliesA | Detects unusual cost spikes across AWS services compared to baseline spending. Use this when asked about billing surprises, unexpected charges, or cost spikes. |
| get_idle_resourcesA | Identifies idle or underutilized AWS resources including EC2 instances, RDS databases, and unattached EBS volumes that are wasting money. Use this when asked about waste, unused resources, or easy cost savings. |
| get_tagging_complianceA | Checks AWS resource tagging compliance — identifying resources missing required tags like Environment, Owner, CostCenter, and Project. Use this when asked about cost allocation, untagged resources, or tagging policies. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Tools are functionally distinct by cloud provider and action, but AWS tools lack a provider prefix (e.g., 'get_cost_summary' vs 'get_azure_cost_summary'), which could cause an agent to misselect if not reading descriptions carefully.
The verb_noun pattern is consistent, but cloud prefixes are applied only to Azure and GCP tools (e.g., 'get_azure_idle_resources') while AWS tools are unprefixed ('get_idle_resources'). This inconsistency makes the naming pattern less predictable.
10 tools is well within the ideal range for a FinOps server. Each tool covers a specific cloud provider and operation (cost summary, idle resources, anomaly detection), making the count appropriate for the domain.
The core FinOps surface (cost, waste, anomalies) is covered for AWS, Azure, and GCP, but tagging compliance is only available for AWS, leaving a notable gap for Azure and GCP users.