Content Core

general MCP Server

Extract what matters from any media source

Install Ready
generalgeneral
2 views159 stars32 forksMIT

Why This Matters

Discovered via unknown and last synced 3mo ago.

Install Ready
Source
unknown
Stars
159
Last synced
3mo ago
Install
Instructions detected

Install

1. Install the package

uvx content-core mcp

2. Add to claude_desktop_config.json

{
  "mcpServers": {
    "content-core": {
      "command": "uvx",
      "args": [
        "content-core",
        "mcp"
      ]
    }
  }
}

Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)

18
Tools
0
Resources
0
Prompts
Standard I/O
Transport

Available Tools (18)

CCORE_AUDIO_CONCURRENCY

Concurrent audio transcriptions (1-10)

CCORE_STT_TIMEOUT

Speech-to-text timeout in seconds

Documents

PDF, DOCX, PPTX, XLSX, EPUB, Markdown, plain text

CCORE_FIRECRAWL_WAIT_FOR

Wait time in ms before extraction

Web

URLs, HTML pages, YouTube videos, Reddit posts

CCORE_FIRECRAWL_PROXY

Firecrawl proxy mode (`auto`, `basic`, `stealth`)

Media

MP3, WAV, M4A, FLAC, OGG (audio); MP4, AVI, MOV, MKV (video)

CCORE_LLM_PROVIDER

LLM provider for summarization

CCORE_URL_ENGINE

URL extraction engine (`auto`, `simple`, `firecrawl`, `jina`, `crawl4ai`)

Category

Formats

FIRECRAWL_API_URL

Custom Firecrawl API URL for self-hosted instances

CCORE_DOCUMENT_ENGINE

Document extraction engine (`auto`, `simple`, `docling`)

CCORE_STT_MODEL

Speech-to-text model

Variable

Description

CCORE_LLM_MODEL

LLM model for summarization

CCORE_STT_PROVIDER

Speech-to-text provider

CRAWL4AI_API_URL

Crawl4AI Docker API URL (omit for local browser mode)

CCORE_YOUTUBE_LANGUAGES

Preferred YouTube transcript languages