Podcast Commerce Intelligence — Claude Code MCP server: setup
Podcast Commerce Intelligence is an MCP server that parses podcast transcripts to find product mentions and sponsorships, format show notes, and track trends across episodes.
no license 1,186 installs updated 2026-03-16
claude mcp add --transport http podcast-commerce-mcp https://podcast-commerce-mcp--sincetoday.run.toolsWhat Podcast Commerce Intelligence does
The server runs over HTTP at https://podcast-commerce-mcp--sincetoday.run.tools. It gives LLM clients access to five tools designed for audio transcript commerce analysis. The extract_podcast_products tool pinpoints items and brands, scoring sentiment strength and sorting categories. Meanwhile, analyze_episode_sponsors evaluates commercial ad segments, identifying CTAs and delivery styles such as host-read or mid-roll spots.
Analysis relies on an episode_id caching layer. Because of this cache, track_product_trends and compare_products_across_shows can benchmark product frequency across multiple releases without reprocessing transcripts. It resolves identical products mentioned across distinct podcasts into a single ranking. To publish results, generate_show_notes_section builds Markdown or HTML lists grouped by endorsement level, appending affiliate links where available.
Cross-show comparisons fail unless you first process transcripts through extract_podcast_products using a persistent episode_id. The pricing model includes 200 complimentary calls daily. Subsequent queries cost $0.01 per call.
What it helps with
- Pull brand and item references from transcript text files
- Classify sponsor segments, delivery formats, and call-to-action hooks
- Assemble shoppable episode descriptions in Markdown or HTML with affiliate tags
- Detect rising and falling product trends across audio releases
Example prompts for the agent
- Extract all products and brand sponsors from transcript
interview-ep12.txt - Build a Markdown show notes section for episode ID
show-101using cached data - Track merchandise trend changes across the last five episodes of the podcast
How to install Podcast Commerce Intelligence
Platform CLI: remote server
claude mcp add --transport http podcast-commerce-mcp https://podcast-commerce-mcp--sincetoday.run.toolsPlatform CLI: remote server
codex mcp add podcast-commerce-intelligence --url https://podcast-commerce-mcp--sincetoday.run.toolsConfig file
{
"mcpServers": {
"podcast-commerce-intelligence": {
"url": "https://podcast-commerce-mcp--sincetoday.run.tools"
}
}
}Config file
{
"servers": {
"podcast-commerce-intelligence": {
"type": "http",
"url": "https://podcast-commerce-mcp--sincetoday.run.tools"
}
}
}Platform CLI: remote server
gemini mcp add -t http podcast-commerce-intelligence https://podcast-commerce-mcp--sincetoday.run.toolsStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about Podcast Commerce Intelligence
How do I install Podcast Commerce Intelligence in Claude Code?
Run claude mcp add --transport http podcast-commerce-intelligence https://podcast-commerce-mcp--sincetoday.run.tools. Claude Code connects directly to the remote HTTP endpoint without local package dependencies.
What are the usage limits and pricing?
The service includes a free tier of 200 calls per day. Beyond that quota, usage costs $0.01 per tool invocation.
Why do trend analysis tools require cached episode IDs?
Functions like track_product_trends and compare_products_across_shows consume pre-extracted data stored under episode_id instead of parsing full audio transcripts each time.