podcast-pipeline

1## Preamble (runs on skill start)
2 
3```bash
4# Version check (silent if up to date)
5python3 telemetry/version_check.py 2>/dev/null || true
6 
7# Telemetry opt-in (first run only, then remembers your choice)
8python3 telemetry/telemetry_init.py 2>/dev/null || true
9```
10 
11> **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`.
12 
13---
14 
15# Podcast-to-Everything Pipeline
16 
17Turns podcast episodes into a full content calendar across every platform.
18One episode in, 15-20 content pieces out — scored, deduplicated, and scheduled.
19 
20---
21 
22## Step 1: Ingest — Get the Transcript
23 
24Determine the input source and obtain a clean transcript.
25 
26### Option A: RSS Feed (`--rss <url>`)
271. Fetch the RSS feed XML
282. Extract the latest episode's audio URL (or use `--episodes N` for batch)
293. Download the audio file
304. Transcribe via OpenAI Whisper API (with timestamps)
315. Store transcript with episode metadata (title, date, description, duration)
32 
33### Option B: Raw Transcript (`--transcript <file>`)
341. Read the transcript file (plain text, SRT, or VTT)
352. Parse timestamps if present
363. Extract episode metadata from filename or prompt user
37 
38### Option C: Batch Mode (`--batch <rss_url> --episodes N`)
391. Fetch RSS feed
402. Extract the last N episodes
413. Process each through the full pipeline
424. Deduplicate across all episodes in the batch
43 
44### Transcript cleanup
45…