discovering-niche-podcasts

Discover niche/industry podcasts, podcast owners, interview episode sources, and guest/interviewee LinkedIn outreach lists using pasted reports, Perplexity, Exa, Tavily, and optional assisted Genspark

SKILL.md

GitHub

discovering-niche-podcasts

Workflow

  1. Confirm the campaign niche, source list, or input file.
  2. Read references/setup.md if the repo is not configured.
  3. Prompt for the minimum intake fields: niche, audience/ICP, optional seed terms, geography, must-include terms, and exclusions.
  4. Run the Apify SERP waterfall first with scripts/run_podcast_intelligence_waterfall.py.
  5. Use YouTube/RSS/transcript/OCR fallback scripts only for high-value shows whose guest names remain missing.
  6. Match both hosts/owners and interviewees to LinkedIn with evidence fields.
  7. Write descriptive outputs under data/ or .tmp/.

Scripts

  • scripts/discover_podcasts.py
  • scripts/ingest_research_report.py
  • scripts/match_linkedin_profiles.py
  • scripts/resolve_youtube_channels.py
  • scripts/scrape_youtube_podcast_guests.py
  • scripts/scrape_audio_podcast_guests.py
  • scripts/transcribe_audio_episodes.py
  • scripts/ocr_thumbnails.py
  • scripts/genspark_assisted.py
  • scripts/extract_youtube_urls_from_report.py
  • scripts/build_interviewee_linkedin_list.py
  • scripts/apify_serp.py
  • scripts/discover_podcast_interviews_serp.py
  • scripts/enrich_linkedin_via_apify_serp.py
  • scripts/run_podcast_intelligence_waterfall.py

References

  • references/research-report-workflows.md for turning pasted reports into podcast, episode, and LinkedIn enrichment outputs.

Guardrails

  • Never hardcode API keys.
  • Never accept low-confidence LinkedIn matches automatically.
  • Do not scrape LinkedIn directly unless the user provides a compliant workflow.
  • Keep raw private research reports out of git unless sanitized.