Most ASO tools want you in their dashboard every morning. But you already spend the day in Claude or Cursor. With MCP (Model Context Protocol), your assistant can read your rankings directly — so “what should I change this week?” becomes one question instead of ten open tabs.
1. Connect once (2 minutes)
Create a key in TrackASO under Settings → API & MCP, then in Claude Code:
claude mcp add --transport http trackaso https://trackaso.app/api/mcp --header "Authorization: Bearer YOUR_KEY"Cursor, Claude Desktop and Codex work the same way — see the setup docs.
2. Give it jobs, not questions
Assistants do best when they know which data to pull and what to hand back. TrackASO ships these as ready-made jobs — in Claude they’re slash commands like /trackaso:weekly_review:
Weekly ASO review
Use TrackASO to review my app “Your App” in the US App Store. Call list_apps to get the app id, then get_rankings and get_weekly_plan. Tell me: 1) the biggest gains and drops since yesterday and where a followed competitor passed me, 2) the 3 searches to focus on this week and why, 3) the exact subtitle (≤30 characters) and keyword field (≤100 bytes) to paste into App Store Connect. Keep it short and concrete.
Write my subtitle and keyword field
For my app “Your App” in the US App Store: get the app id with list_apps, call get_weekly_plan, then optimize_keywords with my name, the suggested subtitle and the plan's searches. Give me 2 subtitle options (≤30 characters each, no words already in my name) and the final keyword field (≤100 bytes, commas, no spaces, no plurals of words already used). Explain each choice in one line.
Find searches my competitors win
For my app “Your App” in the US App Store: call list_apps, then keyword_gap. Pick the 5 searches I can realistically win (low difficulty, real popularity, close to what my app does), check the best two with research_keyword, and offer to start tracking them with add_keywords.
What reviews tell me to fix
For my app “Your App” in the US App Store: call list_apps, then get_reviews. Group the recent reviews into themes (bugs, missing features, praise), say which theme hurts the rating most, and draft a short, friendly reply to the two most critical reviews.
Expand to another country
My app “Your App” is tracked in the US App Store. Using get_rankings, suggest_keywords and research_keyword, pick one more country where it could rank with little competition, list 10 local-language searches to track there, and explain how a second localization (for example Spanish (Mexico) in the US store) adds another 100 bytes of keywords.
3. What a weekly loop looks like
- Monday: run the weekly review. You get what moved, the 3 searches to focus on and a paste-ready subtitle and keyword field.
- Paste them into App Store Connect with your next update.
- TrackASO checks every keyword every night, in the cloud — your Mac can stay off.
- Next Monday, the review shows whether last week’s targets climbed.
Why the data matters more than the model
An assistant without real App Store data guesses: it suggests “habit tracker” to a 20-rating app that will never crack that top 10. With your positions, real popularity and the strength of each search’s top 10, it picks searches you can actually win at your size.