Bruno MCP Server is an MCP Server designed to empower AI models with API testing capabilities using Bruno, a local API client. It acts as a crucial bridge, enabling LLMs to execute Bruno collections and receive detailed, standardized test results.

This server leverages the Bruno CLI to run collections, supporting environment files and variables for flexible testing. Key features include comprehensive test summaries with total, passed, and failed counts, along with detailed failure information and execution timings. AI models can seamlessly interact with Bruno MCP Server using the run-collection tool, providing the collection path and optional environment configurations.

The core value lies in enabling AI to autonomously validate APIs, ensuring reliability and accuracy. Installation is streamlined via Smithery or manual configuration, integrating smoothly with platforms like Claude. By providing a structured interface to Bruno's testing capabilities, Bruno MCP Server significantly enhances the functionality and reliability of AI-driven workflows.

API Test Execution via Bruno

Bruno-mcp allows AI models to execute API tests defined in Bruno, a local API client. It acts as an intermediary, translating the AI model's request into a Bruno CLI command, running the specified Bruno collection, and then parsing the results back into a structured format that the AI model can understand. This enables AI models to validate API functionality, verify responses, and ensure that APIs behave as expected. For example, an AI model could use Bruno-mcp to automatically test a newly deployed API endpoint as part of a CI/CD pipeline, ensuring that no breaking changes are introduced. The server leverages the Bruno CLI for execution and supports environment files and variables for flexible configuration.

Standardized Test Result Interface

Bruno-mcp provides a standardized MCP interface for returning detailed test results to the AI model. This interface includes the overall success/failure status of the test suite, a summary of the number of tests passed and failed, detailed information about any failures, and execution timings. By providing a consistent and well-defined format for test results, Bruno-mcp enables AI models to easily interpret and act upon the results of API tests. For instance, an AI model could use the failure information to automatically identify the root cause of an API issue and suggest potential fixes. The JSON response includes fields for success, summary (total, passed, failed), failures, and timings (started, completed, duration).

Bruno Collection Execution

The core function of Bruno-mcp is to execute Bruno collections, which are sets of API requests and assertions defined using the Bruno API client. The run-collection tool within Bruno-mcp takes the path to a Bruno collection as input and executes all the requests and assertions defined within that collection. This allows AI models to leverage existing API test suites created in Bruno, without needing to understand the underlying details of the API being tested. For example, a developer could create a Bruno collection to test the authentication flow of an application, and then use Bruno-mcp to allow an AI model to automatically verify that the authentication flow is working correctly. The tool also supports optional environment files and variables, allowing for flexible configuration of the test environment.