ai-podcasting

Submit and update AI Podcasting episodes through the scoped WIN client API.

25|1|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/wisdom-in-a-nutshell/agents --skill ai-podcasting-wisdom-in-a-nutshell
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-podcasting
Source: https://github.com/wisdom-in-a-nutshell/agents/tree/main/skills-source/owned/ai-podcasting
Command: npx skills add https://github.com/wisdom-in-a-nutshell/agents --skill ai-podcasting-wisdom-in-a-nutshell

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Managing AI Podcasting episodes normally requires the web GUI, which blocks agent-driven automation. This Skill lets an agent check API access, list TCR episodes with rich metadata, submit new episodes with retry-safe idempotency, and patch show notes, intro copy, titles, and thumbnails directly through the scoped client API. ## Core Features & Use Cases - Access verification and episode listing: Run doctor to validate credentials, show grants, and operation scopes, then list TCR episodes filtered by publication state and date range with rich per-episode summaries. - Retry-safe episode submission: Create episodes via /client/v1/episodes with an Idempotency-Key, preferring Descript web URLs as the main source and rejecting MP3 main sources. - Post-creation copy updates: Patch show notes, asset URLs, and guests on existing episodes, or update intro source, title, thumbnails, and editor instructions on unpublished episodes. - Local file uploads: Upload local media through purpose-scoped S3 cache/ intents when no public URL is available. - Use Case: A producer asks the agent to submit a new TCR episode from a Descript link, then later update its show notes and final thumbnail before publishing, all without opening the web app. ## Quick Start Ask the agent to run the doctor check and list your unpublished TCR episodes, then provide a Descript web URL to submit a new episode.

Frequently Asked Questions about ai-podcasting

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I submit a new AI Podcasting episode from the command line?▼

Run the ai_podcasting_client.py script with the submit-episode subcommand and a JSON payload file containing mainSourceUrl. Prefer a Descript web URL for TCR; MP4 links are accepted as fallbacks while MP3 main sources are rejected.

How do I update show notes on an existing podcast episode?▼

Use the update-episode-copy subcommand with the episode's source ID and a payload containing showNotes, assetUrls, or guests. This is the only supported path for show notes; the intro update command rejects showNotes fields.

Where does the AI Podcasting client store its API key?▼

The client reads the bearer token only from ~/.secrets/aipodcasting/env with mode 600, or from the file path set by AIPODCASTING_CLIENT_API_KEY_FILE. It never accepts the key as a command flag, payload field, or environment value.

What happens if an episode submission fails midway?▼

Submissions send an Idempotency-Key derived from the request ID. Retry the uncertain submission with the same --request-id value; only use a new request ID when you intentionally want a separate episode.

Can I use a local MP4 or image file instead of a public URL?▼

Yes, file-like fields accept local paths, which the aip_local_upload_helper.py script uploads through purpose-scoped intents returning temporary S3 cache URLs. These cache URLs are transport references, not permanent storage.

Why does the intro update command reject my episode?▼

Intro updates target only existing unpublished episodes and require a source ID plus a non-empty payload. Run list-episodes with --publication-state unpublished to find eligible episodes before patching.