R.A.P.I.D.

R.A.P.I.D. is an MCP server that provides local code analysis, project context extraction, and search capabilities across software codebases. Designed for software developers and AI assistants, it enables external MCP clients to understand structural details and inspect code before generating implementations, refactoring files, or producing documentation. The server uses a hybrid architecture featuring a Python MCP server runtime connected via a C FFI layer to a high-performance Rust file scanning library. It scans local project directories to evaluate codebase complexity, parse syntax structures for languages including Python, Rust, C#, and TypeScript/JavaScript, and read or generate plan documentation. Clients can conduct exact text queries with configurable context lines or utilize AI-powered semantic search via machine learning embeddings to locate functions and concepts using plain language queries. By running locally, R.A.P.I.D. gives coding agents deep context while letting developers configure directory depths, file exclusions, and scan timeouts.

Category: Developer Tools & Code Intelligence

Tags: code analysis, search, source-code, static analysis

Visit R.A.P.I.D.

How to install and configure R.A.P.I.D.

1. Clone the repository locally: bash git clone https://github.com/otdavies/Rapid.git cd Rapid 2. Set up and activate a Python virtual environment, then install the dependencies: bash python -m venv venv source venv/bin/activate pip install -r requirements.txt 3. Build the Rust scanner component: bash cd file_scanner cargo build --release cd .. 4. Add the server entry to your MCP client configuration file under mcpServers: json { "mcpServers": { "project-context": { "disabled": false, "timeout": 30, "type": "stdio", "command": "python", "args": [ "/absolute/path/to/Rapid/server.py" ], "env": {} } } }

What you can do with R.A.P.I.D.

• Inspecting a new codebase using initialize_project_context to assess file volume, determine project complexity, and read or generate plan files. * Extracting structured code outlines and hierarchical class or function descriptions across Python, Rust, C#, and JavaScript/TypeScript projects. * Locating specific strings, imports, and variables across entire repositories using exact text search with customizable surrounding context lines. * Finding application logic using natural language prompts via semantic concept search when exact function and variable names are unknown.

Key facts

• Open Source
• https://github.com/otdavies/Rapid
• Developer Tools & Code Intelligence, Web Search & Research
• code analysis, search, source-code, static analysis

Part of MCP Servers

How do I install R.A.P.I.D.?

To install R.A.P.I.D., clone its GitHub repository, establish a Python virtual environment to install dependencies from requirements.txt, and build the Rust scanning library inside the file_scanner folder using cargo build --release. Finally, register the server.py path in your client configuration.

What can R.A.P.I.D. do?

R.A.P.I.D. scans local project directories to provide AI agents with codebase context. It calculates repository complexity, extracts file structures and class or function descriptions, runs project-wide exact text searches, and performs semantic code searches using natural language descriptions.

Which programming languages does R.A.P.I.D. support?

The server includes parsers designed to analyze codebases written in Python, Rust, C#, and TypeScript or JavaScript, in addition to supporting broad textual search across any custom file extensions configured by the user.

Is R.A.P.I.D. open source?

Yes, R.A.P.I.D. is open source software. The full source code, including the Python MCP server and the Rust file scanner library, is available on GitHub.