---
title: "Six bots run a mobile game"
canonical: "https://agenticfeats.com/c/six-bots-run-a-mobile-game-grok-bot/"
agents: ["Grok Bot"]
evidence_grade: "B"
outcome: "completed"
job_type: "Business"
roles: ["Founder or owner", "Product"]
source_url: "https://x.ai/bot/guides/grok-bot-for-mobile-app-development"
source_date: "2026-08-25"
published: "2026-10-13"
updated: "2026-10-13"
---

> From Agentic Feats: sourced examples of AI agents at work. Map of the site for agents: https://agenticfeats.com/llms.txt

# Six bots run a mobile game

The author used six Grok Bot agents with separate roles to run analytics, ads, engineering, deploys, and bug fixes for Rank'em.

Result: Rank'em passed 1000 downloads in a week. The author credits the bots with a 15x lower cost per install and 4x better D7 retention.

## How it was set up

Grok Bot agents with role prompts, cloud computers, nightly routines, recorded click-path skills, and handoffs between bots. The analytics bot connects to Meta Ads, AppLovin, Adjust, PostHog, Sentry, and the app stores.

## Worth knowing

Vendor blog post. Meta Ads API verification failed, so a recorded web UI skill was used. Spend actions need approval.

## Prompt

```text
You are the analytics and user acquisition seat for [app name]. You own paid acquisition. Read what players actually did. Run nightly spend and install recaps across [ad networks], and name winners and losers. Send winners and losers to the creatives bot and product findings to the engineer bot as specs. Ask me before anything that spends money.
```

## Evidence note

Author states 15x CPI and 4x D7 gains; app links, no raw data

## Sources

- Original: https://x.ai/bot/guides/grok-bot-for-mobile-app-development

## The receipt

- Agent: Grok Bot
- Evidence: Numbers (grade B, a specific number stated)
- Outcome: Completed
- Job type: Business
- Roles: Founder or owner, Product
- Source: Blog
- Published: 2026-08-25
- Added: 2026-10-13