add podcast player

[SpkArtZen]

What does this Ability do?

Suggested Trigger Words

Type

• New community Ability
• Improvement to existing Ability
• Bug fix
• Documentation update

External APIs

• No external APIs
• Uses external API(s):

Testing

• Tested in OpenHome Live Editor
• All exit paths tested (said "stop", "exit", etc.)
• Error scenarios tested (API down, bad input, etc.)

Checklist

• Files are in community/my-ability-name/
• main.py follows SDK pattern (extends MatchingCapability, has register_capability + call)
• README.md included with description, suggested triggers, and setup
• resume_normal_flow() called on every exit path
• No print() — using editor_logging_handler
• No hardcoded API keys — using placeholders
• No blocked imports (redis, connection_manager, user_config)
• No asyncio.sleep() or asyncio.create_task() — using session_tasks
• Error handling on all external calls
• Tested in OpenHome Live Editor

Anything else?

[github-actions]

🔀 Branch Merge Check

PR direction: feature/podcast_player → dev

✅ Passed — feature/podcast_player → dev is a valid merge direction

[github-actions]

✅ Ability Validation Passed

📋 Validating: community/podcast-player
  ✅ All checks passed!

[github-actions]

✅ Community PR Path Check — Passed

All changed files are inside the community/ folder. Looks good!

[github-actions]

🔍 Lint Results

✅ __init__.py — Empty as expected

Files linted: community/podcast-player/main.py

✅ Flake8 — Passed

✅ All checks passed!

[abubakar4360]

Hey @SpkArtZen! The idea is great. Just a few things to fix before this can go through:

1. This ability has a textbook menu-driven flow problem. The podcast path runs: speak 3 podcast names one by one → "Choose 1, 2, or 3" → speak 5 episode titles one by one → "Choose 1, 2, 3, 4, or 5" — that's 10 sequential spoken items plus two numbered-choice prompts with no LLM routing between them. Per the QA guide this must go through a manual voice test before approval. Run through it out loud and confirm it doesn't feel like a phone tree. The likely fix is collapsing both levels into a single open-ended prompt — "I found a few podcasts matching that. Which sounds right: The Daily, Crime Junkie, or Serial?" — and letting the LLM match the user's reply to the best option rather than forcing them to say a number.

2. EXIT_WORDS = {"stop", "pause", "exit", "quit", "cancel"} is missing too many natural spoken exits. Add: "forget it", "never mind", "nevermind", "done", "bye", "that's all", "no thanks", "actually", "leave it". "pause" probably shouldn't be in this set at all — a user saying "pause" likely wants to pause playback, not exit the ability entirely.

3. All five word-set checks — SURPRISE_WORDS, SEARCH_WORDS, ELSE_WORDS, WHATS_PLAYING_WORDS, and if "podcast" in text — are brittle in lower substring matches that will miss natural phrasing. A user saying "play something random", "I want to hear a podcast about true crime", "what's on right now", or "got anything else?" will fall through silently. Replace all five with a single LLM intent classifier that returns structured JSON — this ability is simple enough that one classifier handles the whole command space cleanly.

4. The numbered list reading pattern — f"{i}. {ep['title']}" spoken one item per speak() call — produces stiff TTS output. "One. The Daily Monday Episode. Two. Up First from NPR. Three..." sounds like an IVR system. Collapse the list into a single natural sentence: "I found three options: The Daily, Up First from NPR, and Crime Junkie Weekly. Which one?"

5. await self.capability_worker.speak(f"Error streaming audio: {e}") — a raw exception string is going directly to the user's speaker. Change to: "Had trouble loading that episode. Want to try a different one?"

6. "What would you like to listen to? You can search for an episode, a podcast, or say surprise me." — 22 words. The second sentence is a menu read-out that front-loads options the user didn't ask for. Change to: "What do you want to listen to?"

7. "Here are the latest five episodes of {selected_podcast['title']}:" followed by five separate speak() calls — same issue as point 4. Collapse to: "Here are the latest five from {title}: [episode 1], [episode 2]... Which one?" in a single speak call.

8. "No podcasts found." and "No episodes found." are terse to the point of feeling broken. Change to: "Couldn't find any podcasts for that — try a different search." and "That podcast doesn't seem to have any episodes available."

[uzair401]

Hi @SpkArtZen, please address the issues pointed out by @abubakar4360. Once these are resolved, we will proceed further with approval.