podcast-pipeline

Preamble (runs on skill start)

# Version check (silent if up to date)
python3 telemetry/version_check.py 2>/dev/null || true

# Telemetry opt-in (first run only, then remembers your choice)
python3 telemetry/telemetry_init.py 2>/dev/null || true

Privacy: This skill logs usage locally to ~/.ai-marketing-skills/analytics/. Remote telemetry is opt-in only. No code, file paths, or repo content is ever collected. See telemetry/README.md.

---

Podcast-to-Everything Pipeline

Turns podcast episodes into a full content calendar across every platform. One episode in, 15-20 content pieces out — scored, deduplicated, and scheduled.

---

Step 1: Ingest — Get the Transcript

Determine the input source and obtain a clean transcript.

Option A: RSS Feed (--rss <url>)

1. Fetch the RSS feed XML
2. Extract the latest episode's audio URL (or use --episodes N for batch)
3. Download the audio file
4. Transcribe via OpenAI Whisper API (with timestamps)
5. Store transcript with episode metadata (title, date, description, duration)

Option B: Raw Transcript (--transcript <file>)

1. Read the transcript file (plain text, SRT, or VTT)
2. Parse timestamps if present
3. Extract episode metadata from filename or prompt user

Option C: Batch Mode (--batch <rss_url> --episodes N)

1. Fetch RSS feed
2. Extract the last N episodes
3. Process each through the full pipeline
4. Deduplicate across all episodes in the batch

Transcript cleanup

• Remove filler words (um, uh, like, you know) for written content
• Preserve original with timestamps for video clip suggestions
• Split into logical segments by topic shift

---

Step 2: Editorial Brain — Deep Analysis

Feed the full transcript to the LLM with this extraction framework:

Extract these content atoms:

1. Narrative Arcs — Complete story segments with setup → tension → resolution. Tag with start/end timestamps.

2. Quotable Moments — Punchy, shareable statements. One-liners that stand alone. Must pass the "would someone screenshot this?" test.