Rini

MCP server collection

python

2

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mori-mmmm-rini-mcp-server": {
      "command": "python",
      "args": [
        "reasoning.py"
      ]
    }
  }
}

This MCP server collection provides specialized micro-services that generate and execute code, analyze GitHub repositories, perform reasoning, crawl and search the web, summarize YouTube content, and orchestrate MCP servers. You interact with these components via an MCP client to perform targeted data analysis, automation, and code-era tasks in a modular, on-demand fashion.

How to use

You can run each MCP component as a local server and connect to it through an MCP client. Each service exposes a distinct capability, from web search and YouTube analysis to GitHub repository analysis and autonomous reasoning. Start the component you need, send it a query or task, and receive structured results that you can feed into your workflows.

How to install

pip install -r requirements.txt

Prerequisites: ensure Python is installed on your system. The project uses Python-based components that you run directly as scripts.

Start each MCP server individually from its corresponding Python file. Use the exact commands below to run the servers on your machine.

python web_search.py

python youtube_summary.py

python github_repo_analysis.py

python reasoning.py

python coding.py

python web_crawl.py

python mcp_related.py

Additional setup and notes

Each component runs as a local process. If you want to test end-to-end, start multiple components in separate terminals and connect them via your MCP client. The web search component, YouTube processing, GitHub analysis, reasoning, coding, web crawling, and MCP orchestration are designed to operate together when you coordinate their inputs and outputs.

Troubleshooting and tips

If a server does not start, verify that no other process is occupying the required port and that you have the necessary dependencies installed. Review any error messages for missing modules or syntax issues in the respective Python files. Ensure you are running commands from the correct working directory where the Python scripts reside.

Notes on usage and capabilities

This collection includes modules for web search, YouTube content summarization and transcription, GitHub repository analysis, logical reasoning, code generation and execution, text extraction from URLs, and MCP server orchestration. You can combine these capabilities to build complex data processing pipelines, such as fetching web results, analyzing codebases, and generating executable snippets for your projects.

Available tools

rini_google_search_base

Performs a Google search and returns the results.

rini_google_search_link_only

Extracts only the links from Google search results.

rini_google_search_shallow

Performs a Google search and briefly fetches content from each linked page.

rini_summarize_youtube_audio_only

Summarizes only the audio of a YouTube video.

rini_transribe_youtube_audio

Transcribes the audio of a YouTube video into text.

rini_summarize_youtube_all

Analyzes and summarizes the entire YouTube video content across keyframes and audio.

rini_github_analysis

Analyzes a specified GitHub repository, extracting functions/classes and computing code similarity.

rini_reasoning

Performs logical reasoning on a given query using a specified model if provided.

rini_code_generation

Generates code based on a given query, optionally using a specified model.

rini_python_code_execution

Executes provided Python code snippets.

rini_get_text_only_from_url

Extracts text content from a given URL.

rini_get_all_from_url

Fetches all content from a given URL with an optional timeout.

rini_create_and_execute_mcp_server

Generates and executes an MCP server based on a query and filename.