Skip to content
Back to skills

podcast-guest-research

ASecurity

Find podcasts suited to a proposed guest using Crawlora's podcast catalog, live show and episode data, and public show websites. Use for guest-appearance shortlists, recent-topic research, and tailored pitch angles supported by actual episodes.

  • 0 votes
  • 0 copies
  • 1 view
  • Added October 12, 2026
ai-agentsapi

Works with

  • api

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned October 12, 2026

npx -y skills add Crawlora-org/crawlora-skills --skill podcast-guest-research --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of podcast-guest-research?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for podcast-guest-research
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/crawlora-org-podcast-guest-research/badge)](https://www.skillsdirectory.com/skills/crawlora-org-podcast-guest-research)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: podcast-guest-research
description: Find podcasts suited to a proposed guest using Crawlora's podcast catalog, live show and episode data, and public show websites. Use for guest-appearance shortlists, recent-topic research, and tailored pitch angles supported by actual episodes.
---

# Podcast guest research

Find shows where the guest's expertise fits the format and recent editorial
coverage. Produce a sourced shortlist and pitch angles, preserving unknowns about
audience size, booking availability, and whether a show accepts guest proposals.

## Setup

Set `CRAWLORA_API_KEY` to your key from [crawlora.net](https://crawlora.net).
Run the bundled `scripts/crawlora.sh` from this skill directory or by absolute
path. It sends `x-api-key` to `https://api.crawlora.net/api/v1`; keep the key in
the environment. Read [reference/endpoints.md](reference/endpoints.md) for
podcast discovery, live episode lookups, and public website tools.

## Discovery and editorial fit

1. Identify the guest's expertise, credible examples, proposed subjects, language,
   audience, and geographic constraints. Use these to evaluate editorial fit.
2. Discover candidates through `/datasets/apple-podcasts-shows/search` and facets.
   Dataset `q` searches title/artist, not the episode archive. `country` describes
   the discovery storefront, not listener geography or exhaustive availability.
   `sort=popularity` follows chart seed rank; it does not measure downloads.
3. Expand topic discovery with `/apple-podcasts/episodes/search?term=...` or live
   show search. Confirm the correct show/creator and preserve the Apple show ID.
   The stored catalog is sampled from chart/search discovery, not every podcast.
4. Read `/apple-podcasts/show/{id}` and `/show/{id}/episodes` under the same
   `/apple-podcasts` prefix. Inspect recent publication dates, descriptions, and
   guest/interview evidence. Episode lists can be bounded; record what was checked.
5. When useful, resolve the same show using `/spotify-podcasts/search?q=...`.
   Pass a returned `spotify:show:...` URI to `/spotify-podcasts/show` and
   `/spotify-podcasts/show/episodes` (`offset`, `limit` up to 50). Apple numeric
   IDs and Spotify URIs are different identities; match creator/title/website
   before combining records. A show URI is needed even if marked optional.
6. Inspect the official show site or episode notes using `/web/scrape` when
   relevant. Verify booking instructions and public professional contact channels.
   Use short excerpts and distinguish descriptions from a full episode transcript;
   do not claim to have listened to an episode based only on metadata.
7. Propose an angle linked to a specific recent episode or recurring editorial
   theme. Explain the guest's contribution and the new ground it covers. A past
   interview establishes format evidence, not a promise of future bookings.

```sh
scripts/crawlora.sh /datasets/apple-podcasts-shows/search q=founder page_size=5
scripts/crawlora.sh /apple-podcasts/episodes/search \
  term="bootstrapped business" country=us limit=10
```

## Deliverable and bounds

Return show/host, canonical links and IDs, fit rationale, recent episode evidence
with dates, proposed angle, verified submission channel if available, and caveats.
Separate researched contacts from draft pitch copy; send only when requested.

- Deduplicate syndicated shows across providers using verified identity/feed
  evidence. Do not merge by a generic title alone.
- Check recency in the live episode list. A high track count or old chart seed
  does not establish an active show. Preserve missing dates and coverage gaps.
- Rankings, ratings, and episode counts do not establish audience size, audience
  demographics, sponsor rates, or conversion potential.
- Respect dataset `page_size<=100` and its 10,000-result window. Stop on repeated
  pages, exhausted offsets, or the shortlist bound. Back off on `429`, retry a
  transient `5xx` once, stop on `401`/`403`, and check application `code`.

Files in this skill

  • SKILL.md3.9 KB
  • reference/endpoints.md11.1 KB
  • scripts/crawlora.sh6.1 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.