MCP2Lambda

MCP2Lambda is an MCP server that enables Model Context Protocol clients and large language models to discover and execute AWS Lambda functions as standard tools without requiring code changes. It connects AI interfaces such as Claude Desktop or Amazon Bedrock applications to an AWS account, exposing serverless functions as callable actions. Cloud engineers, software developers, and AI application builders use MCP2Lambda to bridge generative models with private infrastructure. The bridge establishes a secure operational boundary through segregation of duties: the client only requires permission to invoke Lambda functions, while the Lambda execution roles manage access to internal VPC networks, private databases, third-party APIs, and other AWS cloud services. The server can operate in pre-discovery mode, where every function matching an allowed prefix is exposed as an individual tool at startup, or generic mode using generic listing and invocation tools. By default, it targets functions prefixed with mcp2lambda-, enabling controlled execution of data lookups, computational workloads, or custom sandboxed Python scripts without exposing backend systems directly to the language model.

Category: Cloud & Infrastructure

Tags: aws, cloud, compute, lambda, serverless

Visit MCP2Lambda

How to install and configure MCP2Lambda

1. Install MCP2Lambda automatically for Claude Desktop using Smithery by running: bash npx -y @smithery/cli install @danilop/MCP2Lambda --client claude 2. For manual setup, clone the repository and configure AWS credentials: bash git clone https://github.com/danilop/MCP2Lambda.git cd MCP2Lambda aws configure 3. Add the server entry to your Claude Desktop configuration file: json { "mcpServers": { "mcp2lambda": { "command": "uv", "args": [ "--directory", "<full path to the mcp2lambda directory>", "run", "main.py" ] } } } 4. Start the server using uv run main.py or restart Claude Desktop.

What you can do with MCP2Lambda

• Query private VPC databases and internal company services securely through scoped Lambda functions without exposing database credentials or networks directly to LLM clients. - Execute sandboxed Python code inside a Lambda environment to calculate complex equations, run prime number computations, or handle data manipulation tasks. - Look up customer records and contact metadata by invoking Lambda functions that interface with customer relationship management backends. - Route AI model requests through Lambda internet access to retrieve real-time external data or interact with authenticated third-party web services.

Key facts

• https://github.com/danilop/MCP2Lambda
• Cloud & Infrastructure, Developer Tools & Code Intelligence
• aws, cloud, compute, lambda, serverless

Part of MCP Servers

What is MCP2Lambda?

MCP2Lambda is a bridge server implementing the Model Context Protocol that allows generative AI models to discover and invoke AWS Lambda functions as tools. It lets models access internal VPC resources, run sandboxed code, or fetch context-specific data without giving the language model direct access to broader AWS infrastructure or private networks.

How do I install MCP2Lambda?

You can install MCP2Lambda automatically for Claude Desktop via Smithery by running npx -y @smithery/cli install @danilop/MCP2Lambda --client claude. You can also manually clone the repository from GitHub, configure your AWS credentials using the AWS CLI, and configure Claude Desktop to run the server script using uv.

Which AWS Lambda functions can MCP2Lambda access?

By default, the server only discovers and exposes Lambda functions with names starting with the mcp2lambda- prefix. It can operate in pre-discovery mode to register each function as an individual named tool at launch, or in generic mode using generic listing and invocation tools to manage interactions.

Which MCP clients work with MCP2Lambda?

MCP2Lambda works with MCP-compliant desktop clients such as Claude Desktop. It also supports custom client applications built on Amazon Bedrock using the Bedrock Converse API, allowing developers to connect Anthropic Claude, Mistral, Llama, or Amazon Nova models directly to Lambda tools.

How does MCP2Lambda handle security and permissions?

The server enforces segregation of duties. The client application only requires AWS credentials with permission to invoke targeted Lambda functions. The Lambda functions then use their own execution roles to access other AWS services, VPC resources, or external networks, preventing the model from having broad AWS permissions.