Skip to content

Usage

CLI reference

The the-data-packet command runs the full pipeline. All options can also be set via environment variables — CLI flags override them.

the-data-packet [OPTIONS]

API keys

Option Env var Description
--anthropic-key KEY ANTHROPIC_API_KEY Anthropic (Claude) API key
--gcs-bucket NAME GCS_BUCKET_NAME GCS bucket for audio synthesis
--google-credentials PATH GOOGLE_APPLICATION_CREDENTIALS Path to GCP service account JSON
--mongodb-username USER MONGODB_USERNAME MongoDB username (optional)
--mongodb-password PASS MONGODB_PASSWORD MongoDB password (optional)
--s3-bucket BUCKET S3_BUCKET_NAME S3 bucket for uploads (optional)

Content

Option Default Description
--sources SOURCE ... wired Sources: wired, techcrunch
--categories CAT ... security ai Categories to fetch
--max-articles N 1 Articles per source

Generation

Option Description
--script-only Generate script, skip audio
--audio-only Generate audio from an existing script

Audio voices

Option Default Description
--male-voice VOICE en-US-Studio-Q Google Cloud TTS voice for first speaker (Alex)
--female-voice VOICE en-US-Studio-O Google Cloud TTS voice for second speaker (Sam)

Available Studio voices

Only en-US-Studio-* voices support multi-speaker Long Audio Synthesis.

Voice ID Character
en-US-Studio-Q Male — warm, professional
en-US-Studio-O Female — clear, engaging
en-US-Studio-M Male — authoritative

Output

Option Default Description
--output DIR ./output Output directory
--show-name NAME The Data Packet Podcast show name
--no-s3 — Disable S3 uploads even if configured
--save-intermediate — Keep intermediate files
--log-level LEVEL INFO DEBUG / INFO / WARNING / ERROR

Common examples

the-data-packet --output ./episode

the-data-packet --script-only --output ./scripts

the-data-packet \
  --show-name "Tech Brief" \
  --male-voice en-US-Studio-Q \
  --female-voice en-US-Studio-O \
  --output ./episode

the-data-packet \
  --sources wired techcrunch \
  --categories security ai \
  --max-articles 2 \
  --output ./multi-source

the-data-packet --no-s3 --output ./local-episode

the-data-packet \
  --log-level DEBUG \
  --save-intermediate \
  --output ./debug

Via Docker

All CLI options work identically when passed to the Docker image:

docker run --rm \
  --env-file .env \
  -v "$(pwd)/output:/app/output" \
  -v "$(pwd)/service-account-key.json:/credentials.json:ro" \
  ghcr.io/thewintershadow/the-data-packet:latest \
  --show-name "Daily Tech Brief" \
  --sources wired techcrunch \
  --categories security ai \
  --max-articles 2

Python API

from the_data_packet import PodcastPipeline, get_config

# With overrides
config = get_config(
    show_name="Tech Brief",
    max_articles_per_source=2,
    article_sources=["wired", "techcrunch"],
    article_categories=["security", "ai"],
    male_voice="en-US-Studio-Q",
    female_voice="en-US-Studio-O",
)

pipeline = PodcastPipeline(config)
result = pipeline.run()

if result.success:
    print(f"Script:   {result.script_path}")
    print(f"Audio:    {result.audio_path}")
    print(f"Time:     {result.execution_time_seconds:.1f}s")
    print(f"Articles: {result.number_of_articles_collected}")
    if result.s3_audio_url:
        print(f"S3 URL:   {result.s3_audio_url}")
else:
    print(f"Failed: {result.error_message}")

Output files

File Description
episode_script.txt Full two-host dialogue with Alex: / Sam: speaker labels
episode.wav Synthesized multi-speaker audio at 44.1 kHz
feed.xml RSS 2.0 podcast feed (only generated when S3 is configured)

Automation

# Daily at 8 AM
0 8 * * * docker run --rm \
  --env-file /opt/podcast/.env \
  -v /opt/podcast/output:/app/output \
  ghcr.io/thewintershadow/the-data-packet:latest

name: Generate Daily Podcast
on:
  schedule:
    - cron: '0 8 * * 1-5'  # Weekdays at 8 AM UTC

jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - name: Generate Podcast
        run: |
          docker run --rm \
            -e ANTHROPIC_API_KEY="${{ secrets.ANTHROPIC_API_KEY }}" \
            -e GCS_BUCKET_NAME="${{ secrets.GCS_BUCKET_NAME }}" \
            -v "/tmp/output:/app/output" \
            ghcr.io/thewintershadow/the-data-packet:latest

      - name: Upload episode
        uses: actions/upload-artifact@v4
        with:
          name: podcast-episode
          path: /tmp/output/

apiVersion: batch/v1
kind: CronJob
metadata:
  name: podcast-generator
spec:
  schedule: "0 8 * * 1-5"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
            - name: podcast-generator
              image: ghcr.io/thewintershadow/the-data-packet:latest
              env:
                - name: ANTHROPIC_API_KEY
                  valueFrom:
                    secretKeyRef:
                      name: podcast-secrets
                      key: anthropic-key
                - name: GCS_BUCKET_NAME
                  valueFrom:
                    secretKeyRef:
                      name: podcast-secrets
                      key: gcs-bucket
          restartPolicy: OnFailure

Troubleshooting

Permission denied on output directory

The container runs as UID 1000. Fix with: 

sudo chown -R 1000:1000 output

API key not found

Verify your env file is being picked up: 

docker run --rm --env-file .env \
  ghcr.io/thewintershadow/the-data-packet:latest \
  --log-level DEBUG

ARM64 platform mismatch

docker pull --platform linux/arm64 \
  ghcr.io/thewintershadow/the-data-packet:latest