Rephonic
OfficialThis server gives AI agents access to Rephonic's podcast database (4+ million shows) for search, research, analysis, and monitoring via MCP tools.
Search & discovery: search podcasts by topic/title/publisher, search episodes by title/notes/transcripts, get autocomplete suggestions, and resolve podcasts from Apple/Spotify/YouTube/RSS IDs.
Podcast insights: look up full podcast metadata, chart rankings, latest episodes, estimated listener counts, demographics, and historical trends.
People & outreach: get hosts/guests, contact emails/pages, social accounts with follower/engagement metrics, and verified contact details.
Content analysis: list/get episode metadata, fetch full transcripts, extract sponsors/promotions with ad copy and promo codes, and read AI-generated review summaries or individual reviews.
Audience & market research: find overlapping/shared audiences, compare podcasts, and analyze listener age, education, profession, income, interests, and location.
Chart & rankings tracking: retrieve Apple, Spotify, and YouTube chart categories/countries/rankings, updated daily.
Monitoring & media lists: track brand mentions, monitor sponsorships, and build targeted media lists at scale.
Filter reference data: list valid categories, countries, languages, sponsors, professions, and interests for advanced search filters.
Usage management: check API quota and usage for the current month.
Allows searching and discovering podcasts using Apple/iTunes IDs, and retrieving chart rankings from Apple Podcasts charts.
Allows searching and discovering podcasts using Spotify IDs, and retrieving chart rankings from Spotify charts.
Allows searching and discovering podcasts using YouTube channel IDs, and retrieving chart rankings from YouTube charts.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Rephonic search for podcasts about AI in business"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Rephonic MCP Server
Give AI agents and LLMs access to podcast data from Rephonic — covering 4+ million podcasts with listener estimates, demographics, contact details, chart rankings, episodes, and more.
This is a remote Model Context Protocol (MCP) server that connects AI assistants like Claude, ChatGPT, and Cursor to the Rephonic API.
Ideas and Use Cases
Read, summarise and analyse transcripts — grab a full-text transcript for almost any podcast episode
Search and discover podcasts — find shows by topic, category, or audience fit
Research podcasts for guest pitching — see listener numbers, audience demographics, and contact details without leaving your AI workflow
Monitor brand mentions — track when your brand, competitors, or clients are mentioned across podcasts
Analyze audience demographics — age, education, profession, income, and location breakdowns for any podcast
Track chart rankings — daily Apple, Spotify, and YouTube chart data across countries and categories
Monitor sponsorships and promotions — see who's advertising on which shows, with ad copy and promo codes
Build media lists — pull verified email contacts, social accounts, and host/guest info at scale
Related MCP server: Sponsorable MCP server
Tools
search_podcasts — Search for podcasts by topic, title, or publisher with advanced filters.
search_episodes — Search for episodes by keyword across titles, show notes, and transcripts.
autocomplete — Get suggested keywords and matching podcasts for a search query.
find_podcast — Resolve a Rephonic podcast from an Apple/iTunes ID, Spotify ID, YouTube channel ID, or RSS feed URL. Use this first when you already have one of those identifiers.
lookup_podcast — Look up a podcast's metadata, chart rankings, and latest episodes by its Rephonic ID.
podcast_people — Get hosts and guests of a podcast with contact details and social profiles.
podcast_demographics — Get estimated listener demographics (age, education, profession, income, location).
podcast_promotions — Get sponsors and cross-promotions from a podcast's episodes.
list_episodes — List all episodes for a podcast in chronological order.
get_episode — Look up metadata for an individual episode.
episode_transcript — Get the full transcript for an episode.
contacts — Get email contacts, contact pages, and social accounts for a podcast.
social_accounts — Get social media accounts linked to a podcast with follower/engagement metrics.
feedback — Get overall ratings and AI-generated review summaries for a podcast.
reviews — Get individual listener reviews from various apps.
trends — Get historical trends for podcast metrics over time.
shared_audience_graph — Get podcasts with overlapping listeners.
chart_categories — List available chart categories and countries for Apple, Spotify, or YouTube.
chart_rankings — Get the latest chart rankings for a platform, country, and category.
list_categories — List podcast categories (for use with search filters).
list_countries — List countries (for use with search filters).
list_languages — List languages (for use with search filters).
list_sponsors — List commonly seen sponsors (for use with search filters).
list_professions — List common listener professions (for use with search filters).
list_interests — List common listener interests (for use with search filters).
api_quota — Check your API request quota and usage.
Setup
You need a Rephonic API key. Get one here.
ChatGPT
Go to Settings → Apps → Advanced settings and enable Developer mode
Then click Create app
Enter Name: Rephonic
Enter MCP Server URL:
https://mcp.rephonic.comClick Create and enter your Rephonic API key when prompted
Claude.ai
Go to Settings → Connectors
Click Add custom connector and enter name
Rephonicand MCP server URLhttps://mcp.rephonic.comClick Add and enter your Rephonic API key when prompted
Claude Desktop
Add to your config file and restart Claude Desktop:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"rephonic": {
"url": "https://mcp.rephonic.com"
}
}
}Claude Code
Add to your project's .mcp.json:
{
"mcpServers": {
"rephonic": {
"type": "http",
"url": "https://mcp.rephonic.com"
}
}
}Cursor
Go to Settings → MCP → Add new MCP server. Set type to URL and enter https://mcp.rephonic.com.
If you want to fork or customize this MCP server for your own needs, here is how you can run or deploy it.
Prerequisites
Node.js
A Cloudflare account (for deployment)
Local dev
npm install
npm startThe server runs at http://localhost:8788. Test with the MCP Inspector:
npx @modelcontextprotocol/inspector@latestType-check
npm run type-checkRegenerate Cloudflare types
After changing wrangler.jsonc:
npm run cf-typegenAvailable Tools
26 toolsapi_quotaAInspect
Check your API request quota and usage for the current month.
Inputs: no named parameters.
autocompleteBInspect
Returns suggested keywords and matching podcasts for a search query.
Inputs: mode (required), query (required).
chart_categoriesAInspect
Get all available chart categories and countries for a platform. Use the returned slugs with the chart_rankings tool.
Inputs: platform (required).
chart_rankingsAInspect
Returns the latest chart rankings for a given platform, country, and category (updated every 24 hours). Use chart_categories to get valid country and category slugs. Use category all for overall top charts.
Inputs: country (required), category (required), platform (required).
contactsAInspect
Returns email contacts, contact pages, and social media accounts for a podcast. Includes quality indicators: concierge (manually verified), warning (likely invalid), and upvotes/downvotes.
Inputs: podcast_id (required).
episode_transcriptCInspect
Get the full transcript for an individual episode. Not available for all episodes. May include speaker name mapping.
Inputs: episode_id (required).
feedbackBInspect
Returns overall rating, total ratings/reviews count across various apps, and AI-generated summary insights of listener reviews.
Inputs: podcast_id (required).
find_podcastAInspect
Find a Rephonic podcast from an external identifier: Apple/iTunes ID, Spotify ID, YouTube channel ID, or RSS feed URL. Use this whenever you already have one of those values — there is no need to call search_podcasts first. Pass exactly one of the four parameters. Returns {podcasts: [...]} — empty when no match; may contain multiple entries when a feed_url or youtube_channel_id is shared. Use the returned id with lookup_podcast and the other podcast tools.
Inputs: feed_url, itunes_id, spotify_id, youtube_channel_id.
get_episodeAInspect
Look up metadata for an individual episode including topics, locations, guests, and memorable moments. Use an ID from the episode search or list_episodes tool.
Inputs: episode_id (required).
list_categoriesAInspect
Returns a list of podcast categories. These IDs are used with the categories search filter.
Inputs: no named parameters.
list_countriesAInspect
Returns a list of countries. These IDs are used with the locations search filter.
Inputs: no named parameters.
list_episodesAInspect
Returns every episode for a given podcast in chronological order, including YouTube videos if found.
Inputs: podcast_id (required), page, query, per_page.
list_interestsAInspect
Returns common listener interests, optionally filtered by name. These names are used with the interests search filter.
Inputs: query.
list_languagesAInspect
Returns a list of languages. These codes are used with the languages search filter.
Inputs: no named parameters.
list_professionsAInspect
Returns common listener professions, optionally filtered by name. These names are used with the professions search filter.
Inputs: query.
list_sponsorsAInspect
Returns commonly seen sponsors, optionally filtered by name. These names are used with the sponsors search filter.
Inputs: query.
lookup_podcastAInspect
Look up a podcast's metadata, chart rankings, and latest episodes by its Rephonic ID (e.g. 'huberman-lab'). If you have an Apple/iTunes ID, Spotify ID, YouTube channel ID, or RSS feed URL instead of a Rephonic ID, use find_podcast first to resolve it. If you only know the name, use search_podcasts.
Inputs: podcast_id (required).
podcast_demographicsAInspect
Returns estimated demographic information about a podcast's listeners including age groups, education levels, professions, interests, household income, and country-level geographic distribution.
Inputs: podcast_id (required).
podcast_peopleBInspect
Returns hosts and guests of a podcast with names, descriptions, affiliations, contact emails, contact pages, and social media profiles. Data is extracted from transcripts and may have occasional inaccuracies.
Inputs: podcast_id (required).
podcast_promotionsBInspect
Returns sponsors and cross-promotions from a podcast's episodes. May include product/service details, full ad read text, call to action, URLs, and promo codes. Data is extracted from transcripts and may have occasional inaccuracies.
Inputs: podcast_id (required).
reviewsAInspect
Returns individual listener reviews from various apps in chronological order.
Inputs: platform (required), podcast_id (required).
search_episodesAInspect
Search for episodes. Matches against episode titles, show notes, and full transcripts. Specify either query or filters or both.
Inputs: page, query, filters, per_page, +3 more.
search_podcastsAInspect
Search for podcasts by topic, title, or publisher with advanced filters. Returns estimated listeners per episode in the downloads_per_episode field. Specify either query or filters or both. Filters are comma-separated, e.g. listeners:gte:5000,active:is:true. See list_categories, list_countries, list_languages, list_sponsors, list_professions, list_interests for valid filter values.
Inputs: mode, page, query, filters, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 26 tools
Most tools have distinct purposes, but list_episodes and lookup_podcast both provide episode data, causing minor overlap. podcast_people, contacts, and social_accounts also overlap in providing contact info, though differentiated by detail type.
Predominantly follows verb_noun pattern with underscores, e.g., search_podcasts, lookup_podcast. A few tools like 'feedback' and 'contacts' are just nouns, deviating slightly from the convention.
26 tools is above the typical range but justified for a comprehensive podcast data API covering search, metadata, demographics, contacts, reviews, charts, and trends. Each tool serves a specific function without redundancy.
The tool set covers the full spectrum of podcast data: discovery, metadata, episodes, transcripts, people, demographics, promotions, contacts, social, feedback, reviews, trends, shared audience, and charts. No obvious gaps for a read-only data API.
Related MCP Connectors
Podcast search, metadata, chapters, and transcripts for AI agents — from $15/mo
Podcast knowledge for AI agents: search episodes, pull transcripts and ask questions.
Podcast intelligence for agents: transcripts, clips, speaker diarization, mention tracking.
Real SEO data for AI assistants: page audits, Keyword Planner volumes, Search Console history.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceWraps the Podcast Index API (podcastindex.org) to enable AI agents to search and retrieve podcast episodes and metadata.297 npmMIT
- AlicenseNot gradedqualityCmaintenanceConnects a podcast-sponsorship database to AI assistants, allowing users to search for brands that sponsor podcasts and reveal buyer contacts by name and email.MIT
- AlicenseAqualityBmaintenanceEnables AI agents to discover long-tail podcasts through randomized probe queries, trend lookups, and RSS feed peeks, providing cadence, episode length, and feed details without requiring API keys.6MIT
- AlicenseAqualityBmaintenanceEnables AI agents and Claude Code to query podcast download analytics from OP3, including unique listeners, retention cohorts, audience geography and apps, episode performance, and show comparisons via natural language.2234 npmMIT
social_accountsAInspect
Returns social media accounts linked to a podcast (Instagram, Facebook, X/Twitter, Patreon, TikTok, LinkedIn, etc.) with follower counts and engagement metrics.
Inputs: podcast_id (required).