Podcast Workflow

Build a complete podcast production system via OpenClaw Ultra. From episode planning and script generation to AI audio cleanup, social clip creation, and multi-platform distribution, manage your entire podcast operation from a single chat interface.

Core System Overview

INFO

This is a closed-loop podcast production workflow. OpenClaw Ultra handles episode planning, script drafting, AI audio cleanup guidance, transcript review, content repurposing, clip creation, and performance analysis, so you can publish consistently without spending every weekend editing.

System Layer	Core Function	Final Output
Planning Layer	Topic research, episode outlining, guest research, content calendar	Scheduled episodes with outlines ready to record
Script Layer	Episode script generation, intro/outro writing, segment structuring	Recording-ready scripts in natural speaking language
Recording Layer	Recording setup guidance, guest briefing, prep checklists	Clean recording session with minimal technical issues
Cleanup Layer	AI audio editing guidance, filler word removal, noise reduction, leveling	Polished audio ready for post-production
Transcript Layer	Automated transcription, show notes, chapter markers, episode description	Searchable transcript and episode metadata
Repurposing Layer	Content extraction, blog post generation, newsletter copy, audiogram scripting	20+ content assets from one recording
Clips Layer	AI clip detection, short-form clip extraction, caption generation	Platform-ready social clips for TikTok, Reels, Shorts, LinkedIn
Distribution Layer	RSS publishing, platform submission, social scheduling, performance tracking	Live episode across all platforms with promotion assets
Review Layer	Episode performance analysis, audience feedback synthesis, workflow tuning	Data-driven improvements for next episode

Prerequisites

Item	Requirement
OpenClaw Ultra	Installed and running
Recording Setup	Microphone, headphones, quiet recording space
Audio Hosting Account	(Recommended) Riverside, Descript, Podbean, Buzzsprout, or Transistor for RSS hosting
Transcription Service	(Recommended) Riverside (free unlimited), Otter.ai, or Descript
Clip Repurposing Tool	(Recommended) Opus Clip, Kapwing, or Descript for AI clip generation
Email Marketing Tool	(Recommended) ConvertKit, Mailchimp, or Beehiiv for newsletter distribution

Step 0 — Initialize Your Podcast Production System

Set up OpenClaw Ultra as your podcast production manager.

Operation Steps

1. Open a new chat session on OpenClaw Ultra
2. Have your podcast concept, existing episode list, and recording workflow ready
3. Paste the initialization prompt
4. For content repurposing beyond podcast audio, see the SEO Content Workflow

Ready-to-Use Prompt

Act as my podcast production manager.

My show:
- Podcast name: [name]
- Format: [solo / interview / co-host / hybrid]
- Niche/topic: [brief description]
- Target audience: [who listens and why]
- Current episode count: [number]
- Publishing cadence: [weekly / biweekly / monthly]
- Average episode length: [minutes]

My biggest production problems:
- I spend [6-12] hours on post-production editing per episode, mostly removing filler words and dead air
- I never have time to create social clips — my episodes get zero promotion
- Show notes and episode descriptions are an afterthought, if I write them at all
- I have [dozens / hundreds] of past episodes that never got repurposed into clips or blog posts
- My content calendar runs dry every few weeks and I scramble for episode ideas
- Transcription takes forever and I never review it for accuracy

Build me a production system that:
- Plans episodes weeks in advance so I never scramble for ideas
- Generates episode outlines and scripts that sound like natural speech, not written prose
- Guides me through AI-powered audio cleanup so editing takes minutes instead of hours
- Creates a transcript that's accurate enough to publish as show notes and a blog post
- Extracts 15-30 social clips per episode automatically
- Turns every recording into blog posts, newsletter copy, quote cards, and LinkedIn posts
- Tracks episode performance so I know what topics resonate and what to do next

Step 1 — Plan Your Episode Calendar

Replace the last-minute scramble with a planned content pipeline that generates episodes weeks ahead.

1.1 Generate Episode Ideas for the Next 8 Weeks

Prompt

Generate 8 episode topics for my podcast: [podcast name], about [niche], for [target audience].

For each episode, provide:
- Episode title (3 options — punchy, searchable, shareable)
- Core thesis (one sentence: what will listeners learn or feel after this episode)
- 3 main talking points
- Ideal format (solo deep-dive / interview with a specific guest type / panel / case study)
- Suggested guest if applicable (role + the angle they'd bring)
- Hook (one compelling line to use in episode intro or promo)

Rules:
- Vary the formats so not every episode is the same type
- Include 1-2 episodes that address common misconceptions in your niche
- Include 1 episode tied to a current event or trend in [industry]
- Distribute difficulty: some should be evergreen, some timely
- Match the publishing cadence of [weekly / biweekly] in mind

Return as a table: Episode #, Title Option, Format, Guest (if any), Core Talking Points (3 bullets), Hook.

1.2 Create a Deep-Dive Episode Outline

Prompt

Build a detailed outline for an episode about: [topic from your calendar].

Format: [solo / interview / panel]
Target length: [30-45] minutes
Audience level: [beginner / intermediate / advanced]

Structure requirements:
- Opening hook (first 60 seconds — grab attention immediately)
- Context setting (why this topic matters right now, 1-2 minutes)
- Main body: [4-6] segments, each with a specific point, evidence, and transition
- Each segment should be [5-8] minutes
- Q&A or listener questions segment (if applicable)
- Closing summary (3 key takeaways)
- Call to action (what should listeners do after this episode — subscribe, rate, visit site, etc.)

For each segment:
- Segment name and purpose
- Key points to hit (bullet format)
- Story or example to include
- Transition line to next segment

Rules:
- Outline should be skimmable — I need to see the structure at a glance during recording
- Include specific questions if this is an interview episode
- Note where I should pause for listener engagement (questions to ask the audience)
- Flag any segments that overlap with recent episodes so you can avoid repetition

Step 2 — Draft Your Episode Script

Write scripts that sound like natural conversation, not written essays.

2.1 Generate a Full Episode Script

Prompt

Write a complete episode script for my podcast about: [topic].

Format: [solo / interview]
Target length: [30-45] minutes
Tone: [conversational / authoritative / interview-style / storytelling]

Write the script in natural spoken language — this will be read aloud, not read silently.

Rules:
- Use short sentences. Spoken English runs 130-160 words per minute; a 40-minute episode needs [5,200-6,400] words
- Avoid passive voice and academic phrasing
- Include placeholders for: [story or example you'd share here], [data point to verify], [transition phrase]
- For interview episodes: write the host questions in full, write brief framing for guest answers (let the guest bring the substance)
- Add [PAUSE] markers before key transitions
- Add [SOUND EFFECTS] or [MUSIC] cues where appropriate
- Don't over-script the intro — make it feel natural, not read from a teleprompter

Format the output with clear section headers so I can navigate quickly during recording.

Return the full script with word count per section. Target word count: [X].

2.2 Write Episode Intro and Outro

Prompt

Write the intro and outro for my episode about: [topic].

Intro requirements:
- Duration: [30-60] seconds
- Hook: open with a striking fact, question, or statement that makes someone listen to the end of the intro
- Name drop: mention the show name and this episode's focus clearly
- Guest intro (if applicable): [X] sentence bio that establishes credibility
- Preview: one sentence on what listeners will get from this episode
- Tone: energetic but not fake — conversational confidence

Outro requirements:
- Duration: [45-90] seconds
- Recap: three key takeaways in one sentence each
- Call to action: subscribe, rate, review, share (pick the one that matters most right now)
- Teaser: hint at next episode topic or a reason to come back
- Sign-off: consistent closing phrase I'll use every episode

Return as two separate scripts with timing estimates. The intro should work standalone if someone finds you mid-episode and needs context.

Step 3 — Guide AI Audio Cleanup

Walk through your post-production workflow using AI tools, from raw recording to polished episode.

3.1 Create Your Cleanup Checklist by File

Prompt

I've just exported the raw audio from my recording session. The file is: [filename / URL].

Walk me through a step-by-step cleanup in this order:

Step 1 — Backup and organize
- Save the raw file as an archive before touching anything
- Create a project folder: [podcast name] - Episode [XX] - [YYYY-MM-DD]
- Name the raw file: raw_[episode number].wav

Step 2 — Noise reduction
- Use [Descript / Adobe Podcast Enhance Speech / Auphonic / Cleanvoice] to run a noise reduction pass
- Settings to use: [your preferred tool and settings]
- Listen to 30 seconds before and after to check for artifacts (robotic tinny quality = too aggressive)

Step 3 — Filler word removal
- Run filler word detection: um, uh, like, you know, basically, actually, I mean, so yeah
- Do NOT remove every instance — only remove the ones that disrupt the flow or repeat within [5] seconds
- Review each flagged section before approving removal
- Target: reduce filler word density by [60-70]%, not [100]%

Step 4 — Dead air and silence trimming
- Remove silences longer than [2] seconds
- Check for accidental long pauses mid-sentence — trim those
- Don't remove all silence — natural speech needs breathing room

Step 5 — Audio leveling
- Normalize loudness to [-16 LUFS] for podcast standard
- Check for volume spikes (guest laughing, coughing) and reduce to match
- Run a final pass with [Auphonic / Lumen5] for consistent leveling

Step 6 — Intro/outro insertion
- Add intro music (file: [path]) at [0:00], fade in over [2] seconds
- Add outro music (file: [path]) starting at [timestamp of end], fade out over [3] seconds
- Include sponsor read placement marker if applicable

Return a checklist I can tick off as I go through each step. Note which tools to use and where to find them.

3.2 Flag Audio Issues for Human Review

Prompt

Review the cleaned audio for [episode name] and flag any issues that need human attention.

Check for:
1. Sentences that got cut awkwardly by AI filler word removal
2. Places where the audio cuts feel abrupt (should have a soft transition)
3. Cross-talk moments where both speakers talk over each other
4. Audio glitches: clipping, dropouts, echo
5. Names or numbers that the AI might have transcribed incorrectly
6. The first and last [30] seconds of the episode — these need extra quality review

Return a table: timestamp, issue type, what I hear, recommended fix. For anything that could just be a weird quirk of the recording (not an actual problem), mark it as "keep as-is."

Rules:
- Flag real problems only. If it sounds fine, say so.
- For cross-talk: decide if both voices matter or if one should be reduced
- For name mis-transcriptions: provide the correct spelling

Step 4 — Generate Transcripts and Show Notes

Turn your audio into searchable, publishable written content.

4.1 Generate a Full Episode Transcript

Prompt

Generate the transcript for [episode name], episode [number] of [podcast name].

[episode description or outline]

Rules:
- Speaker labels for each speaker change
- Timestamps every [2-3] minutes for easy navigation
- Mark [laughter], [pause], [music], [sound effect] in brackets
- Include the intro and outro scripts verbatim
- For interview episodes: attribute questions to the host, responses to the guest by name
- Verify spelling of all names and technical terms — check against your existing content if you have it
- Mark any segments with content that should NOT be published (personal info, mistakes, off-topic tangents) with [DO NOT PUBLISH]

Format: plain text with speaker labels. Return the full transcript ready for review and editing.

4.2 Create Show Notes and Episode Metadata

Prompt

Based on the transcript, create full episode metadata for [episode name].

Required output:
1. Episode title: clear, searchable, under [60] characters
2. Episode description: [150-250] words, conversational tone, includes key topics and guest info, ends with a CTA
3. 5 key takeaways: one sentence each, what listeners will learn
4. Chapter markers: list the main segments with [timestamp] and [title] (aim for [5-7] chapters)
5. Guest links: social media handles and website for each guest
6. Resources mentioned: any tools, books, links, or people discussed in the episode
7. Timestamp outline: brief summary of what happens at each chapter

Rules:
- The description must include [2-3] natural keywords for discoverability — don't keyword stuff
- Chapter titles should describe the topic, not just label it (e.g., "Why Most Pricing Pages Fail" not "Chapter 3")
- Include the full transcript as an appendix for SEO
- The description is what shows on Apple Podcasts and Spotify — make it compelling enough that someone scrolling past will tap play

Step 5 — Repurpose Content Across Channels

Extract maximum value from every recording by generating multiple content assets.

5.1 Turn One Episode into a Blog Post

Prompt

Transform the transcript from my episode about [topic] into a long-form blog post.

Target: [1,500-2,500] words, optimized for SEO

Structure:
- SEO title (under 60 chars) and meta description (under 155 chars)
- Introduction (hook the reader the same way the episode opens)
- Key sections that mirror the episode chapters
- Pull quotes from the transcript (2-3 powerful statements)
- A "Lessons Learned" or "Key Takeaways" summary section
- CTA at the end (listen to the full episode, subscribe, etc.)
- Proper heading hierarchy (H2/H3)

Rules:
- Write for someone who will read, not just listen — expand on things the audio implies
- Include the transcript timestamps as links within the post
- Add alt text to any embedded images: [descriptive alt text including topic keywords]
- Keep paragraphs short — 2-4 sentences max
- End with a question to drive comments

5.2 Generate Newsletter Content and Social Copy

Prompt

Create promotional content from my episode about [topic] for these channels:

1. Email newsletter segment ([100-150] words):
- Hook that makes someone open the email
- Brief episode summary
- One memorable quote from the episode
- Listen link with a compelling reason to tune in now
- Signature with podcast artwork

2. LinkedIn post ([150-200] words):
- Open with the most surprising or counterintuitive claim from the episode
- Share the top [2] insights in short bullet points
- End with a question that drives comments
- Include the episode link in the first comment

3. Twitter/X thread ([3-5] tweets):
- Tweet 1: hook (the most compelling one-liner from the episode)
- Tweets 2-4: key insights, one per tweet, with a number
- Tweet 5: CTA + link

4. Quote cards (3 quotes):
- Pull 3 quotable moments from the transcript
- For each, suggest an accompanying visual direction (e.g., "stat on a dark background," "white text on photo of [guest]")

Rules:
- Match the tone of each platform — LinkedIn is more professional, Twitter is punchier
- Don't repeat the same content across channels — adapt it
- Quote cards should be under [280] characters including attribution

Step 6 — Create Short-Form Clips at Scale

Extract 15-30 social clips from every episode without scrubbing through the audio manually.

6.1 Identify Clip Opportunities from the Transcript

Prompt

Analyze the transcript from [episode name] and identify the best clip candidates for short-form social media.

Clip criteria — score each potential clip on:
1. Hook strength (first 3 seconds must grab attention without episode context)
2. Self-contained value (does this clip make sense if the viewer hasn't heard the episode?)
3. Emotional resonance (laugh, surprise, agreement, inspiration — passive listening isn't shareable)
4. Length fit (target [30-90] seconds for TikTok/Reels, [60-120] seconds for YouTube Shorts)
5. Shareability (does this make someone want to send it to a friend?)

Prioritize:
- Strong opinions with clear reasoning
- Specific data points or statistics
- Contrarian takes or surprising claims
- Actionable frameworks or steps
- Story arcs with a clear payoff
- Moments where the host or guest visibly reacts (surprise, laughter, emphasis)

Return the top [15-20] clip candidates as a table: clip #, timestamp, duration, hook line (first 10 words), why it scores high, recommended platform (TikTok / Instagram Reels / YouTube Shorts / LinkedIn / Twitter).

Exclude: context-dependent jokes that won't land without the episode, filler segments, transitions, anything that requires visual context to make sense.

6.2 Write Clip Captions and Hooks

Prompt

Write captions and hooks for these [number] clip candidates from [episode name]:

[list of clip timestamps and topics]

For each clip:
1. Opening hook (first 2-3 seconds): write the exact first line that will appear on screen
2. Caption text: [200-300] characters for TikTok/Reels, [125-150] for LinkedIn
3. Hashtag set: [5-7] hashtags mixing broad reach (#podcast #business) with niche-specific
4. Comment bait: one question at the end of the caption to drive comments and boost algorithmic reach
5. CTA: follow, listen to the full episode, or share

Rules:
- Hooks must work with captions off (most viewers watch without sound)
- Vary the caption styles across clips — don't use the same format for every one
- Captions should tease the insight without giving away the full clip value
- Hashtags should match the clip's actual content, not just the podcast name

6.3 Generate Audiogram Scripts

Prompt

Write audiogram scripts for [number] audio clips from [episode name].

An audiogram is a short audio clip with a waveform visual — for platforms where video isn't practical.

For each clip:
- Duration: [30-60] seconds
- Edit the transcript to stand alone: add brief context so the audio makes sense without the episode
- Write a [1-2] sentence intro that frames the clip
- Write a [1] sentence outro that drives to the full episode

Audiogram visual direction:
- Background: [solid color / branded gradient / podcast artwork]
- Waveform style: [animated / static]
- Text overlay: [hook line / quote / topic label]
- Speaker name and podcast logo: [position and style]

Rules:
- Audio must work without visuals (people listen to audiograms on mute too — make the transcript compelling on its own)
- Keep intros under [10] words
- Match the visual style to your podcast branding consistently
- Return as a brief script + visual direction note for each clip

Step 7 — Publish and Track Episode Performance

Distribute your episode across all platforms and monitor what resonates.

7.1 Generate a Publishing Checklist by Platform

Prompt

Create a publishing checklist for episode [number]: [episode title].

Platforms to publish on:
- Apple Podcasts
- Spotify
- [YouTube / other platforms]
- My website (RSS-hosted)
- [Newsletter platform]
- Social media schedule

For each platform, specify:
1. What to upload (audio file, video file, audiogram, transcript)
2. Metadata required (title, description, tags, category, explicit tag, episode number, season)
3. Cover art specs (dimensions, file format, whether to update show art or use episode-specific art)
4. Scheduling: publish immediately or schedule for [date/time]
5. Any platform-specific requirements (Spotify has specific requirements for marked/explicit episodes)
6. RSS submission or directory listing step if this is a new directory

Rules:
- Double-check episode number and season in the RSS feed — incorrect numbering breaks listener subscriptions
- Submit to Apple Podcasts at least [48] hours before you want it live — Apple reviews new episodes
- For YouTube: upload both the full video episode and the top [3] clips
- Schedule social posts to go out [2-4] hours after the episode publishes

Return as a checklist that I can work through in order, with timestamps for scheduled posts.

7.2 Review Episode Performance and Adjust

Prompt

Review the performance data for episode [number]: [episode title].

Data to analyze:
- Downloads/streams by platform: Apple Podcasts, Spotify, YouTube, direct RSS
- Completion rate (did people listen to the whole episode or drop off at [X] minute?)
- Clip views: [number] views on each social platform
- Website traffic from episode show notes (sessions, average time on page, bounce rate)
- Newsletter open/click rate if applicable
- Social engagement: comments, shares, saves per clip

Questions to answer:
1. Which segment had the highest listener retention (check the chapter skip data if available)?
2. Which clip performed best — and why? (hook style, topic, platform, timing?)
3. Where did listeners come from? (Organic search? Social referral? Direct subscription?)
4. Did the episode meet the production goal? (If the goal was growth: did it bring new subscribers? If the goal was engagement: did it drive comments?)

Return a summary table: metric, this episode, prior 3-episode average, trend, and a one-sentence interpretation.

Based on the performance, suggest [2-3] adjustments for the next episode — specific, actionable, not generic ("improve quality").

Rules:
- Compare to your prior [3] episodes for context, not to viral benchmarks
- Flag anything that needs immediate action (low completion rate = topic or pacing issue; high traffic but no downloads = website not converting)

Final Closed-Loop Podcast Workflow

Topic Researched → Outlined → Scripted → Recorded → Audio Cleaned → Transcribed → Show Notes Written → Clips Identified → Clips Created → Episode Published → Social Clips Distributed → Performance Reviewed → Insights Applied → Next Topic Researched

Practical Usage Tips

1. Record with video enabled even if your primary audience is audio-only. YouTube is now the top platform for monthly podcast consumption for [33]% of US listeners (New Media, 2025), and talking-head clips consistently outperform audiogram clips on TikTok and Instagram in both completion rate and engagement.
2. Batch your recording sessions. Record [3-4] episodes in one sitting when motivation is high, then edit and publish on a schedule throughout the month. Podcast Insights' 2026 survey found that batching is the single most effective scheduling change for maintaining consistency without burnout.
3. Set your AI filler word removal to [60-70]% reduction, not 100%. Over-cleaned audio sounds robotic. Listeners expect natural speech patterns — the goal is removing the jarring stuff, not eliminating every "um."
4. Extract clips within [48] hours of publishing. Clip tools like Opus Clip report peak algorithmic evaluation in the first [24-48] hours after posting — clips published late miss their discovery window.
5. Use timestamps in show notes as navigation hooks. Listeners who land mid-episode from search are much more likely to stay and listen from a timestamp link than from the beginning.
6. For repurposing podcast content into written SEO content, see the SEO Content Workflow — a well-structured blog post from your episode transcript can rank for months and drive ongoing discovery.
7. For connecting your podcast performance to a broader content analytics overview, see the Content Calendar Workflow.