---

name: openclaw-podcast description: Transform your OpenClaw workspace into personalized AI-powered podcast briefings. Get daily audio updates on your work, priorities, and strategy in 8 compelling styles—from documentary narrator to strategic advisor. Connects directly to your agent's memory and files. Includes 4 free hours of podcast generation with your Superlore.ai API key. Schedule morning, midday, or evening briefings that keep you informed without reading screens.

openclaw-podcast

Generate personalized daily podcast briefings from your OpenClaw agent's memory.

Your AI agent already knows what you did today — your memory files, project status, metrics, and decisions. This skill turns that context into a professional podcast briefing via Superlore.ai. Zero manual input required.

Includes 4 free hours of podcast generation with your API key.

Setup

1. Sign up at superlore.ai
2. Go to Account → API Keys and create a new API key
3. Set the environment variable:

   export SUPERLORE_API_KEY=your-api-key-here
   

   Add this to your ~/.zshrc or ~/.bashrc to persist across sessions.

Quick Start

# Generate a briefing with the default style
node scripts/generate-episode.js

# Try different styles
node scripts/generate-episode.js --style "The Advisor"
node scripts/generate-episode.js --style "10X Thinking"

# Preview the prompt without creating an episode
node scripts/generate-episode.js --dry-run

# Schedule automatic briefings
node scripts/setup-crons.js

First Episode in 60 Seconds

Already have your API key? Here's the fastest path to your first briefing:

export SUPERLORE_API_KEY=your-key-here
node scripts/generate-episode.js --dry-run    # Preview what gets sent
node scripts/generate-episode.js              # Generate your first episode!

Your episode will be ready in ~2 minutes at the URL printed in the output. That's it — your agent just turned today's work into a podcast.

What It Does

The skill reads your OpenClaw workspace files:

• memory/YYYY-MM-DD.md — Daily activity logs
• JOBS.md — Project status and job board
• HEARTBEAT.md — Current priorities and upcoming work
• MEMORY.md — Long-term context (optional)

It then pre-processes this data into a structured briefing (accomplishments, metrics, blockers, upcoming work) and generates a professional podcast episode using Superlore's API.

Privacy & Data Flow

Your workspace contains sensitive information. Here's exactly how this skill protects it:

What happens when you generate a briefing

Your Machine (private)              Superlore API (remote)
┌─────────────────────┐             ┌──────────────────┐
│ 1. Agent reads your │             │                  │
│    workspace files  │             │ 4. GPT writes    │
│                     │             │    podcast script │
│ 2. Script extracts  │             │                  │
│    work summaries   │   ONLY      │ 5. Kokoro TTS    │
│                     │  ──────►    │    generates      │
│ 3. Secrets stripped │  briefing   │    audio          │
│    (keys, emails,   │   text      │                  │
│    IPs, paths, DBs) │             │ 6. Private episode│
│                     │             │    stored         │
└─────────────────────┘             └──────────────────┘

What IS sent

• A processed briefing summary (~3,000 chars of work narrative)
• Style instructions (e.g., "be a documentary narrator")
• Voice/speed preferences

What is NEVER sent

• ❌ Your raw workspace files
• ❌ API keys, tokens, passwords, or credentials
• ❌ Email addresses
• ❌ IP addresses or internal hostnames
• ❌ Database connection strings
• ❌ File paths from your system
• ❌ SSH credentials
• ❌ Agent configuration (model routing, cron schedules, etc.)

How secrets are stripped

The script applies three layers of sanitization:

1. Section filtering — Entire sections of MEMORY.md that contain operational details (credentials, model routing, cron configs) are excluded before any data extraction.

2. Line filtering — Individual lines mentioning API keys, passwords, tokens, database URLs, or service IDs are dropped.

3. Pattern stripping — A final regex pass on the entire briefing catches anything that looks like a secret: long tokens (sk_*, ghp_*, rpa_*), database URLs (postgres://), email addresses, internal IPs, base64 blobs, and absolute file paths.

Verify it yourself

Run --dry-run to see exactly what would be sent to Superlore — nothing hidden:

node scripts/generate-episode.js --dry-run

This prints the complete prompt. If you see anything sensitive, report it so we can improve the sanitization rules.

Episode visibility

All briefing episodes are created with visibility: private. This is hardcoded in the script and cannot be overridden by flags or configuration. Only you can access your briefings.

What Superlore sees

Superlore's API receives the briefing text, processes it through GPT (for script writing) and Kokoro TTS (for audio). The episode is stored on Superlore's infrastructure as a private episode tied to your account. Superlore does not have access to your filesystem, your OpenClaw agent, or any files beyond the briefing text you send.

Briefing Styles

Choose from 8 compelling perspectives—each designed to give you a different lens on your work:

1. The Briefing — Your personal documentary. A professional narrator walks through your day's accomplishments, key metrics, obstacles, and what's coming next. Clear, structured, and comprehensive.

2. Opportunities & Tactics — Spot what you're missing. This style hunts for growth opportunities, strategic angles, and tactical moves hiding in your daily work. Perfect when you need fresh ideas.

3. 10X Thinking — Break through limits. A moonshot advisor challenges your assumptions, questions your approach, and pushes you to think 10x bigger. Use when you feel stuck in incremental thinking.

4. The Advisor — Honest feedback from a seasoned mentor. No sugar-coating—just direct guidance on what's working, what's not, and what to do about it. Your trusted counsel.

5. Focus & Priorities — Cut through the noise. Ruthlessly identifies what actually matters and what's just busywork. Helps you say no to distractions and double down on leverage.

6. Growth & Scale — All about the numbers. Revenue, users, conversion funnels, and growth loops. Analyzes your work through the lens of business metrics and scalability.

7. Week in Review — Step back and see patterns. Synthesizes your week into trends, lessons learned, and clear goals for the week ahead. Best on Friday or Sunday evenings.

8. The Futurist — Connect today to tomorrow. Ties your daily work to your 3, 6, and 12-month vision. Keeps you aligned with long-term goals while moving fast.

See references/styles.md for full style definitions and prompt templates.

Voices

Three curated voices are available for your briefings:

| Voice | Character | Best For | |-------|-----------|----------| | Luna (af_heart-80_af_sarah-15_af_nicole-5) | Warm, balanced — female | The Briefing, The Advisor, Week in Review | | Michael (am_michael-60_am_eric-40) | Rich, resonant — male blend | Opportunities, 10X Thinking, Growth & Scale | | Heart (af_heart) | Soft, intimate — female | Focus & Priorities, The Futurist |

Luna is Superlore's signature voice — a carefully balanced blend that sounds natural across a wide range of content. Michael brings authority to strategic and growth-focused styles. Heart is the most personal of the three, ideal for styles that speak directly to the individual.

Custom Styles

Not limited to the 8 defaults! Create your own podcast style by dropping a JSON file in your workspace's podcast-styles/ directory:

// podcast-styles/my-style.json
{
  "name": "My Custom Style",
  "description": "Brief description",
  "voice": "af_heart-80_af_sarah-15_af_nicole-5",
  "speed": 0.95,
  "targetMinutes": 6,
  "instructions": "Detailed instructions for the AI narrator..."
}

Voice options for custom styles:

• "af_heart-80_af_sarah-15_af_nicole-5" — Luna (warm, balanced)
• "am_michael-60_am_eric-40" — Michael (rich, authoritative)
• "af_heart" — Heart (soft, intimate)

Then use it:

node scripts/generate-episode.js --style "My Custom Style"
node scripts/generate-episode.js --custom "my-style"  # matches filename
node scripts/generate-episode.js --list-styles  # see all available

Two example custom styles are included: Founder Debrief (casual co-founder conversation) and Competitor Watch (competitive intelligence lens).

Scheduling

Use setup-crons.js to configure automatic daily briefings:

node scripts/setup-crons.js

The setup wizard guides you through 7 steps:

1. API Key Validation — Test your Superlore key before anything else; shows remaining hours
2. Choose your styles — Pick from the 8 built-in styles (e.g., "The Briefing" mornings, "The Advisor" evenings)
3. Custom style — Optionally design your own briefing with a custom focus, tone, and voice
4. Music bed — Choose ambient background music or voice-only
5. Schedule — Set a time for each style (morning/midday/evening/weekly/custom cron)
6. Preview episode — Generates a real episode so you can hear the quality immediately
7. Activate — Outputs openclaw cron add commands; optionally runs them for you

Each built-in style has a curated default voice:

• Luna voices: The Briefing, The Advisor, Week in Review
• Michael voices: Opportunities & Tactics, 10X Thinking, Growth & Scale
• Heart voices: Focus & Priorities, The Futurist

Example output:

# The script will generate commands like:
/cron create "0 8 * * *" agent:run node scripts/generate-episode.js --style "The Briefing" --time-of-day morning

# Just copy-paste these into your OpenClaw chat to activate

Once scheduled, your briefings will generate automatically and be ready to listen at your chosen times. You'll receive notifications with episode links.

Configuration

The script uses sensible defaults but accepts these parameters:

• --style <name> — Briefing style (default: "The Briefing")
• --time-of-day <morning|midday|evening> — Affects greeting and tone
• --date YYYY-MM-DD — Generate for a specific date (default: today)
• --dry-run — Preview exactly what would be sent (nothing hidden)
• --no-memory — Skip MEMORY.md entirely (only use daily files)
• --api-url <url> — Override Superlore API URL
• --device-id <id> — Optional device identifier for usage analytics (not required for authentication)
• --channel <channel> — Deliver the episode link to a specific channel (e.g., telegram, discord) when the episode is ready

Requirements

• OpenClaw workspace with memory files
• Superlore.ai API key (get yours at superlore.ai)
• Internet connection (episodes created via Superlore API)
• 4 free hours of podcast generation included with your API key

Learn More

• Superlore.ai — https://superlore.ai
• API Reference — references/api.md
• Style Guide — references/styles.md

Tips

• Start with "The Briefing" style to get a feel for the format
• Use "The Advisor" when you want honest feedback on your work
• Try "10X Thinking" when you feel stuck or want to challenge assumptions
• "Week in Review" works best on Friday or Sunday evenings
• Use --dry-run to experiment with prompts before creating episodes
• Morning briefings focus on the day ahead; evening briefings review accomplishments

Examples

# Morning briefing with advisor style
node scripts/generate-episode.js --style "The Advisor" --time-of-day morning

# Weekly review for Friday
node scripts/generate-episode.js --style "Week in Review" --time-of-day evening

# Preview what a growth-focused briefing would look like
node scripts/generate-episode.js --style "Growth & Scale" --dry-run

# Generate and deliver the episode link to Telegram when ready
node scripts/generate-episode.js --channel telegram

---

Made with ❤️ for the OpenClaw community. Powered by Superlore.ai.