Nable (finops Mcp)
@chaandannn
关于 Nable (finops Mcp)
Local-first FinOps copilot: ask about your AWS, Azure, GCP, and SaaS costs in plain English. Anomaly detection, rightsizing, idle-resource cleanup, and Jira/Linear ticketing, all in your own Claude. Credentials never leave your machine.
基本信息
配置
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"nable": {
"command": "uvx",
"args": [
"finops-mcp"
]
}
}
}工具
未检测到工具
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
概览
What is Nable (finops MCP)?
Nable (finops MCP) is a local-first FinOps copilot that lets users ask about AWS, Azure, GCP, and SaaS costs in plain English inside Claude. It runs entirely on your own machine with anomaly detection, rightsizing, idle-resource cleanup, and Jira/Linear ticketing—credentials and cost data never leave your machine.
How to use Nable (finops MCP)?
Install by adding the provided MCP server configuration to your Claude desktop app, or run pip install finops-mcp. Once configured, ask natural-language questions about cloud costs (e.g., "What drove our AWS costs up 40% this month?") and the server returns actionable insights.
Key features of Nable (finops MCP)
- Cost queries across AWS, Azure, GCP, and 13+ SaaS providers
- Anomaly detection with built-in alerting
- Rightsizing and idle-resource cleanup recommendations
- Cost attribution by team and service
- Commitment (Savings Plans and Reserved) recommendations
- Jira, Linear, and GitHub ticketing integration
Use cases of Nable (finops MCP)
- Investigate a sudden AWS cost spike (e.g., 40% month-over-month increase)
- Identify which team or service is driving the highest Datadog spending
- Surface EC2 rightsizing opportunities to reduce waste
- Discover idle resources wasting money across cloud accounts
- Auto-create a Jira ticket for any cost anomaly over $500
FAQ from Nable (finops MCP)
What makes Nable (finops MCP) different from other FinOps tools?
Nable is local-first: all credentials and cost data remain on your machine, never sent to external servers. It runs as an MCP server inside Claude for a conversational interface.
Which cloud providers and SaaS services does it support?
It supports AWS, Azure, GCP, and over 13 SaaS providers (e.g., Datadog).
Where is my cost data stored?
Your cost data and cloud credentials never leave your machine. The tool runs locally and communicates only with the cloud/SaaS APIs you authorize.
Can I create tickets from anomalies or recommendations?
Yes. Nable can create tickets in Jira, Linear, and GitHub for any detected anomaly or recommendation.
What runtime does it require?
It requires Python and the uvx tool (or a direct pip install). The MCP server configuration uses uvx with the finops-mcp package.
云与基础设施 分类下的更多 MCP 服务器
Mcp K8s Go
strowkMCP server connecting to Kubernetes
Supabase MCP Server
coleam00Supabase MCP server created in Python.
AWS Model Context Protocol (MCP) Server
alexei-ledA lightweight service that enables AI assistants to execute AWS CLI commands (in safe containerized environment) through the Model Context Protocol (MCP). Bridges Claude, Cursor, and other MCP-aware AI tools with AWS CLI for enhanced cloud infrastructure management.
GCP MCP
eniayomiA Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with your Google Cloud Platform environment. This allows for natural language querying and management of your GCP resources during conversations.
AWS MCP Servers
awslabsOpen source MCP Servers for AWS
评论