Podcast Inventory Muse Task Card — Mechanical Steps for Tier-0/1 Agents

At a glancePodcast Inventory Muse Task Card — Mechanical Steps for Tier-0/1 Agents
  1. 1What This Card Is For
  2. 2Prerequisites
  3. 3Phase 1 — YouTube Channel Inventory

What This Card Is For

This is the mechanical portion of the podcast inventory — the tier-0/1 work that does not require frontier-model judgment. A lightweight agent (Muse, Qwen, Gemini Flash) can run every step here. When you hit a judgment call — entity disambiguation, scoring, strategic prioritization — stop, document what you found, and hand the workbook to the frontier model.

SCOPE — WHAT YOU DO AND DON’T DO
You do: platform searches, data extraction, CSV export, sheet population, basic dedup (exact URL match), link verification, count reconciliation.
You don’t: guest identification from ambiguous names, entity disambiguation (two “John Smith” entries), voice/tone writing, scoring or ranking, strategic decisions about what to repurpose first.

Prerequisites

  • A Google Sheet with five tabs already created: Playlists, All Episodes, Summary, Listen Notes, Podchaser. (If the sheet doesn’t exist yet, create it with these tabs before starting.)
  • The subject’s name, exactly as spelled on their profiles.
  • URLs for: their YouTube channel, Listen Notes profile (if known), Podchaser profile (if known).

Phase 1 — YouTube Channel Inventory

Trigger: You have a YouTube channel URL.

  1. Navigate to the channel’s Videos tab. Sort by date (oldest first).
  2. For each video over 10 minutes (likely a full episode or interview):
    • Record in the All Episodes tab: Title, URL, Upload Date, Duration, View Count, Description (first 200 chars).
  3. Check the channel’s Playlists tab. For each playlist that groups podcast episodes or interview series:
    • Record in the Playlists tab: Playlist Name, URL, Episode Count, Date Range.
  4. Run a YouTube search: "[subject name]" podcast OR interview. Add any results hosted on OTHER channels where the subject is a guest — mark these as “Guest Appearance” in the Type column.
  5. Count check: Total episodes in All Episodes tab should match the channel’s public video count (minus shorts, clips, and non-podcast content). Record the delta and reason in the Summary tab.

Phase 2 — Listen Notes Search

Trigger: Phase 1 complete, or you’re running Listen Notes independently.

  1. Search Listen Notes for the subject’s exact name in quotes.
  2. Export results (Listen Notes allows CSV export on paid plans; otherwise extract manually).
  3. For each result, record in the Listen Notes tab: Episode Title, Podcast Name, Air Date, Duration, Listen Notes URL.
  4. Basic dedup against the All Episodes tab: match by exact episode title or URL. Mark matches as “Also on YouTube” in a Crossref column. Do NOT attempt fuzzy matching — flag near-matches for the frontier model.
  5. Record in the Summary tab: total Listen Notes results, how many matched YouTube, how many are unique to Listen Notes.

Phase 3 — Podchaser Credits

Trigger: Phase 2 complete, or you’re running Podchaser independently.

  1. Search Podchaser for the subject’s name.
  2. If a profile exists, record in the Podchaser tab: each credited appearance with Episode Title, Podcast Name, Role (Host/Guest), Air Date, Podchaser URL.
  3. If no profile exists, note “No Podchaser profile found” in the Summary tab and move on.
  4. Basic dedup against All Episodes and Listen Notes tabs by exact title match. Mark crossrefs.
  5. Record in Summary: total Podchaser credits, crossref counts.

Phase 4 — Summary and Handoff

  1. Fill the Summary tab with:
    • Total unique episodes across all platforms (after exact dedup).
    • Platform breakdown: YouTube-only, Listen Notes-only, Podchaser-only, multi-platform.
    • Date range: earliest and latest episode dates.
    • A “Needs Judgment” section listing every row where you weren’t sure — fuzzy title matches, ambiguous guest names, episodes that might be the same content under different titles.
  2. Hand the completed workbook to the frontier model with this note: “Mechanical inventory complete. [N] episodes across [M] platforms. [X] rows flagged for judgment in the Summary tab’s Needs Judgment section.”
COMPLETION CHECKLIST
  1. All Episodes tab has every video >10 min from the YouTube channel
  2. Guest appearances from other channels are included and marked
  3. Listen Notes results are recorded with crossrefs to YouTube
  4. Podchaser credits are recorded with crossrefs
  5. Summary tab has platform counts that add up
  6. Every ambiguous row is flagged in “Needs Judgment” — you did not guess
  7. No entity disambiguation was attempted — names are recorded as-found
  8. The handoff note states the exact counts and the flag count
FLEET ROUTING NOTE
This card is designed for tier-0/1 agents (Muse, local Qwen, Gemini Flash). If you are a frontier model (Claude Opus, GPT-4, Gemini Ultra), you should run the full YouTube SOP or the full cross-platform SOP instead — they include the judgment steps this card deliberately omits.

Parent hub: How We Turned Four Podcast Inventory Drafts into Agent-Ready SOPs

Related: How We Inventoried 700+ Podcast Appearances (Meta Article)

Originally published .

Dennis Yu
Dennis Yu
Dennis Yu is the CEO of Local Service Spotlight, a platform that amplifies the reputations of contractors and local service businesses using the Content Factory process. He is a former search engine engineer who has spent a billion dollars on Google and Facebook ads for Nike, Quiznos, Ashley Furniture, Red Bull, State Farm, and other brands. Dennis has achieved 25% of his goal of creating a million digital marketing jobs by partnering with universities, professional organizations, and agencies. Through Local Service Spotlight, he teaches the Dollar a Day strategy and Content Factory training to help local service businesses enhance their existing local reputation and make the phone ring. Dennis coaches young adult agency owners serving plumbers, AC technicians, landscapers, roofers, electricians, and believes there should be a standard in measuring local marketing efforts, much like doctors and plumbers must be certified. He has appeared on 353 podcasts with 619 credited episodes — see the full list of his podcast appearances.