abanoub-ashraf/indie-aso-metadata-automation

Indie ASO Metadata Automation

Overview

Use this skill to run a practical pipeline:

1. Fetch ASA search-term performance directly from API (or use CSV).
2. Rank keywords by popularity + quality signals.
3. Compose App Store metadata suggestions.
4. Generate (or apply) asc commands to populate metadata quickly.

Use this workflow when the user asks for:

• ASO keyword recommendations backed by ASA data.
• Metadata automation in App Store Connect.
• Fast update loops for title/subtitle/keywords/description.

Preconditions

Require:

• asc installed and authenticated (asc auth ... already done).
• App context from user: app positioning, target audience, brand name, and constraints.

For live ASA fetch (run without --input or fetch command), require one auth mode:

• --asa-access-token (or ASA_ACCESS_TOKEN).
• --asa-client-id + --asa-client-secret.
• --asa-client-id + --asa-key-file + --asa-team-id + --asa-key-id.

Workflow

1) One-command end-to-end (recommended)

Run:

python3 .codex/skills/indie-aso-metadata-automation/scripts/aso_metadata_pipeline_v2.py run \
  --start-date 2026-01-01 \
  --end-date 2026-01-31 \
  --org-id "YOUR_ASA_ORG_ID" \
  --asa-client-id "YOUR_CLIENT_ID" \
  --asa-key-file "/path/AuthKey_ABC123.p8" \
  --asa-team-id "YOUR_TEAM_ID" \
  --asa-key-id "ABC123XYZ" \
  --brand "YourBrand" \
  --app-id "1234567890"

2) CSV mode (fallback)

Use this when API creds are not available.

Run:

python3 .codex/skills/indie-aso-metadata-automation/scripts/aso_metadata_pipeline_v2.py run \
  --input ./asa-search-terms.csv \
  --start-date 2026-01-01 \
  --end-date 2026-01-31 \
  --brand "YourBrand" \
  --app-id "1234567890"

3) Step-by-step mode

Rank:

python3 .codex/skills/indie-aso-metadata-automation/scripts/aso_metadata_pipeline_v2.py rank \
  --input ./asa-search-terms.csv \
  --output ./asa-ranked.csv \
  --top 25

Compose:

python3 .codex/skills/indie-aso-metadata-automation/scripts/aso_metadata_pipeline_v2.py compose \
  --ranked ./asa-ranked.csv \
  --brand "YourBrand" \
  --output ./metadata.json

If needed, include:

• --primary "forced keyword"
• --description-file ./description.txt
• --whats-new-file ./whats-new.txt

Generate asc automation commands:

Run:

python3 .codex/skills/indie-aso-metadata-automation/scripts/aso_metadata_pipeline_v2.py asc \
  --metadata ./metadata.json \
  --app-id "1234567890" \
  --locale "en-US"

This prints commands without executing. Use --apply only after user confirmation.

4) Optional title/subtitle patch via localizations file

First download exact app-info localization file from ASC:

asc localizations download --app "1234567890" --type app-info --locale "en-US" --path ./app-info.en-US.strings

Then run:

python3 .codex/skills/indie-aso-metadata-automation/scripts/aso_metadata_pipeline_v2.py asc \
  --metadata ./metadata.json \
  --app-id "1234567890" \
  --locale "en-US" \
  --app-info-strings ./app-info.en-US.strings

This patches name/subtitle in the strings file and adds the upload command.

Operating rules

• Always show proposed metadata and commands before running --apply.
• Keep App Store limits strict:
  • Title <= 30
  • Subtitle <= 30
  • Keyword field <= 100
• Prefer ASA-backed terms over guesswork.
• Preserve brand constraints and avoid keyword stuffing.

Resources

• Script (legacy): scripts/aso_metadata_pipeline.py
• Script (current): scripts/aso_metadata_pipeline_v2.py
• Expanded examples and troubleshooting: references/workflow.md