More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Meeza Marketing: game UA & retention analysis

    1

    Server and docs: https://github.com/azeemullah67/meeza-marketing-mcp Mobile game marketing lives in four dashboards that disagree with each other.

    Free

    Security scanned
    Meeza Marketing: game UA & retention analysis

    Meeza Marketing: game UA & retention analysis

    Example session with this skill installed

    How did Puzzle Quest's paid installs do over the last two weeks, and should I raise the budget?

    • Read your context and instructions
    • Compiled the meeza marketing

    Summary

    Puzzle Quest, Oct 26 – Nov 8 (14 days) vs the previous 14 days. Sources: Google Ads + Meta Ads for spend and installs, GA4 for first opens and retention. Demo project with synthetic numbers.

    KPIThis periodPreviousChange
    Ad spendUSD 2,643USD 2,556+3.4%
    Installs (GA4 first_open)4,7414,705+0.8%
    Blended CPIUSD 0.78USD 0.76+2.5% (target USD 1.80, 230% of target)
    D1 retention31.2%30.7%+0.5 pt (3,684 users, 11 mature cohorts)
    D7 retention11.8%11.6%+0.2 pt (1,685 users, 5 mature cohorts)

    Verdict: scale carefully. Retention and cost both support growth. Add budget in steps and re-check CPI and D7 after each step.

    Alerts (2)

    High CPI: PQ | US | Installs | UAC - CPI USD 1.12 vs blended 0.78. 24% of spend. Lower its budget or bid target, or move budget to the LATAM campaign (CPI 0.41).
    2.

    High CPI: PQ | US | Video prospecting - CPI 1.05, frequency 3.6 on Meta: the audience is saturating. Refresh creatives before adding budget.

    Creatives

    AdPlatformSpendInstallsCPIStage
    Video - Satisfying mergesMeta6121,0900.56scale
    Video - Boss levelMeta3387300.46scale
    App Ad #92002Google Ads3497020.50steady
    Video - Level 42 failMeta4764021.18cut
    Static - Combo x5Meta2251711.32cut

    Next steps

    • Move ~USD 25/day from "Level 42 fail" and "Combo x5" into "Satisfying merges" and "Boss level".
    • Hold US UAC budget flat until its 7-day CPI is under USD 0.90.
    • Revenue is not tracked in GA4 yet, so ROAS and LTV are estimates only. Link AdMob to Firebase to make them real.

    Data read: get_project_overview, get_marketing_alerts, get_creative_test_board, get_project_retention (all read-only).

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Free and open source. Server and docs: https://github.com/azeemullah67/meeza-marketing-mcp

    Mobile game marketing lives in four dashboards that disagree with each other. This skill teaches your agent to use the Meeza Marketing MCP server, which reads your own Meta Ads, Google Ads, Google Play Console and Google Analytics 4 / Firebase accounts, so you can ask questions in plain language and get numbers with their source, date range and caveats.

    What it does

    • UA performance: spend, installs, CPI, CPM, CTR, IPM by platform, campaign, ad, asset and country, with period-over-period change.
    • Retention: D1/D3/D7/D30 cohorts with maturity, blended curves, per-segment comparison and a soft-launch go / no-go scorecard.
    • Player behaviour: level progression funnel (where players quit), tutorial completion, DAU and stickiness.
    • Creative decisions: a testing board that sorts every ad into scale / steady / refresh / cut with the reason.
    • Store health: Play reviews, ratings, crash and ANR rates next to the marketing numbers.
    • Attribution sanity: Meta installs, Google installs, GA4 first opens and store installs side by side, never summed.
    • Dashboard: one command opens a local web dashboard with the same data.

    How the skill helps the agent

    It knows which of the 50+ tools answers which question, defaults to sensible date ranges, and applies the rules that keep analysis honest: never add installs across sources, keep currencies per account, treat small cohorts as inconclusive, do not read an unmatured cohort as a drop, say when revenue or win rates are not tracked.

    Requirements

    • The Meeza Marketing MCP server running locally (.NET 9, Windows, macOS or Linux). Install steps in the repository README.
    • Read-only credentials for the platforms you use: Meta System User token (ads_read), Google Ads service account or OAuth, GA4 service account with Viewer access, Play Console service account. Credentials stay in an encrypted store on your machine.

    Safety

    Every tool is read-only and tested to stay that way. Nothing can change a campaign, budget or bid. Optional allow-lists limit which accounts the server may read.

    Known limitations

    • Needs the MCP server configured first; setup takes about 20 minutes per platform.
    • Revenue, ROAS and LTV require revenue events in GA4 (for example AdMob linked to Firebase).
    • Store-listing visitors need the Play Console reports bucket.

    Read-only. Cannot edit budgets or pause ads. Requires local MCP server setup and platform API tokens. Data accuracy depends on GA4 custom metric configuration.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 8 days ago

    • Free with an Agensi account

    Requires

    Meeza Marketing (GitHub: Azeemullah67/Meeza Marketing Mcp) MCP
    .NET 9 runtime
    Read-only credentials for Meta Ads, Google Ads, GA4 and/or Google Play Console

    Permissions

    Network Access

    The skill is instructions only. All data access happens in the MCP server running on the user's own machine; it only reads (GET / report queries), stores credentials in an encrypted local store, and honours account allow-lists.

    Listed8 days ago

    What's inside

    Frequently Asked Questions

    The skill aggregates live data from Meta Ads, Google Ads, GA4, and the Google Play Console, allowing your agent to identify performance discrepancies and calculate blended ROAS without manual dashboard hopping.

    This skill is designed for any AI agent or platform that supports the Model Context Protocol (MCP), enabling seamless communication between your LLM and the Meeza Marketing server.

    Setup involves connecting the skill to the Meeza Marketing MCP server and providing authorized API credentials for your specific ad and analytics accounts.

    For security and stability, the skill has read-only access, meaning it can analyze and report on your data but cannot modify budgets, pause campaigns, or change account settings.

    Yes, the skill includes specialized tools for cohort analysis (D1, D7, D30) and can filter level progression data by app version to help you pinpoint where players are churning.

    New in Marketing & SEO