polibase-mcp-server

wimjan123/polibase-mcp-server

3.1

If you are the rightful owner of polibase-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

This Model Context Protocol (MCP) server provides access to a political transcript database, allowing users to retrieve quotes, search transcripts, filter by speakers/politicians, and analyze political discourse data.

Tools
6
Resources
0
Prompts
0

Political Transcript Database MCP Server

This Model Context Protocol (MCP) server provides access to a political transcript database, allowing you to retrieve quotes, search transcripts, filter by speakers/politicians, and analyze political discourse data.

Features

  • Quote Retrieval: Get political quotes with flexible filtering options
  • Date/Time Search: Search quotes within specific date ranges and time periods
  • Speaker Analysis: Filter quotes by specific politicians or speakers
  • Full-Text Search: Advanced search through transcript content
  • Database Statistics: Get insights about the database contents
  • Rich Metadata: Access sentiment analysis, readability scores, and more

Installation

  1. Clone or copy the MCP server files
  2. Install dependencies:
npm install
  1. Build the TypeScript code:
npm run build
  1. Configure environment variables (copy .env.example to .env and adjust as needed):
cp .env.example .env

Configuration

Set the following environment variables:

POSTGRES_HOST=localhost
POSTGRES_PORT=5432
POSTGRES_DB=political_transcripts
POSTGRES_USER=postgres
POSTGRES_PASSWORD=postgres

Available Tools

1. get-quotes

Retrieve political quotes from transcript segments with optional filtering.

Parameters:

  • limit (optional): Maximum number of quotes to return (default: 50, max: 1000)
  • speaker (optional): Filter by speaker name
  • source (optional): Filter by video source
  • date_from (optional): Filter from date (YYYY-MM-DD format)
  • date_to (optional): Filter to date (YYYY-MM-DD format)
  • min_words (optional): Minimum word count for quotes
  • sort_by (optional): Sort criteria - "date", "speaker", "relevance", "word_count" (default: "date")

Example:

{
  "limit": 10,
  "speaker": "Trump",
  "date_from": "2024-01-01",
  "sort_by": "word_count"
}

2. search-quotes-by-date

Search for quotes within specific date ranges and time periods.

Parameters:

  • date_from (required): Start date (YYYY-MM-DD format)
  • date_to (optional): End date (defaults to same as start date)
  • time_from (optional): Start time within video (HH:MM:SS format)
  • time_to (optional): End time within video (HH:MM:SS format)
  • speaker (optional): Filter by specific speaker
  • limit (optional): Maximum results (default: 100, max: 500)

Example:

{
  "date_from": "2024-03-15",
  "time_from": "14:30:00",
  "time_to": "15:00:00",
  "speaker": "Biden"
}

3. get-quotes-by-speaker

Retrieve quotes from specific speakers or politicians.

Parameters:

  • speaker (required): Speaker/politician name to search for
  • party (optional): Filter by political party
  • limit (optional): Maximum results (default: 100, max: 500)
  • min_words (optional): Minimum word count per quote
  • date_from (optional): Filter from date (YYYY-MM-DD)
  • date_to (optional): Filter to date (YYYY-MM-DD)
  • sort_by (optional): Sort order - "date", "word_count", "relevance"

Example:

{
  "speaker": "Trump",
  "party": "Republican",
  "min_words": 50,
  "sort_by": "word_count"
}

4. search-transcripts

Perform full-text search through transcript content with advanced filtering.

Parameters:

  • query (required): Search query text
  • search_type (optional): Type of search - "fulltext", "exact", "fuzzy" (default: "fulltext")
  • speaker (optional): Filter by speaker name
  • source (optional): Filter by video source
  • date_from (optional): Filter from date (YYYY-MM-DD)
  • date_to (optional): Filter to date (YYYY-MM-DD)
  • sentiment (optional): Filter by sentiment - "positive", "negative", "neutral"
  • min_words (optional): Minimum word count
  • limit (optional): Maximum results (default: 50, max: 500)
  • sort_by (optional): Sort order - "relevance", "date", "speaker", "sentiment"

Example:

{
  "query": "climate change",
  "search_type": "fulltext",
  "sentiment": "positive",
  "date_from": "2024-01-01",
  "sort_by": "relevance"
}

5. get-speakers

Get list of available speakers/politicians in the database.

Parameters:

  • limit (optional): Maximum speakers to return (default: 100, max: 1000)
  • party (optional): Filter by political party
  • min_segments (optional): Minimum number of transcript segments

Example:

{
  "party": "Democrat",
  "min_segments": 10
}

6. get-database-stats

Get statistics about the political transcript database.

Parameters: None

This tool returns comprehensive database statistics including:

  • Total videos, segments, speakers, and topics
  • Date range of available content
  • Word count statistics
  • Source distribution

Usage with Claude Code

To use this MCP server with Claude Code, add it to your MCP configuration:

{
  "mcpServers": {
    "polibase": {
      "command": "node",
      "args": ["/path/to/polibase-mcp/dist/index.js"],
      "env": {
        "POSTGRES_HOST": "localhost",
        "POSTGRES_PORT": "5432",
        "POSTGRES_DB": "political_transcripts",
        "POSTGRES_USER": "postgres",
        "POSTGRES_PASSWORD": "postgres"
      }
    }
  }
}

Development

Running in Development Mode

npm run dev

Building

npm run build

Linting

npm run lint

Database Schema

The MCP server works with the following main database tables:

  • videos: Video metadata including title, date, source, and statistics
  • transcript_segments: Individual transcript segments with speaker information, text, timestamps, and analytics
  • speakers: Speaker/politician information with aggregated statistics
  • topics: Topic classifications for content analysis

Each transcript segment includes:

  • Speaker information and political party
  • Full transcript text with word counts
  • Video timestamps and duration
  • Sentiment analysis scores (multiple algorithms)
  • Readability metrics
  • Content moderation flags
  • Stress analysis scores

Error Handling

All tools include comprehensive error handling and will return descriptive error messages if:

  • Database connection fails
  • Invalid parameters are provided
  • Queries fail or timeout
  • No results are found

Security

  • Uses parameterized queries to prevent SQL injection
  • Validates all input parameters with Zod schemas
  • Supports SSL database connections in production
  • Environment variable configuration for sensitive data

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Add tests if applicable
  5. Submit a pull request

License

MIT License