What are ASO.dev Agent Skills?

They are reusable SKILL.md instruction folders that teach AI agents how to handle App Store Optimization, app marketing, metadata, keywords, reviews, monetization, and ASO.dev workflows.

Do Agent Skills replace ASO.dev?

No. Skills guide the AI assistant. ASO.dev remains the product interface, data source, local Companion surface, and final place where drafts are reviewed.

Do I need MCP to use the skills?

No. Skills can guide manual analysis and drafts. The local AI Companion MCP makes them more useful because an agent can read current ASO.dev state and fill drafts when you explicitly ask.

Is the local MCP the same as the public API?

No. The localhost AI Companion MCP and the ASO.dev Developer API are separate surfaces with different contracts, auth, and release status.

ASO.dev Agent Skills: AI Workflows for ASO and App Marketing

In Development Coming Soon

ASO.dev Agent Skills are ready-made AI workflows for App Store Optimization and app marketing. They help AI agents work with ASO.dev more predictably: read the right context, draft only the requested fields, validate changes, and fill drafts only when you ask.

The skills live in the public GitHub repository:

ASO.dev Agent Skills on GitHub

When to Use Skills

Use Agent Skills when you want an AI assistant to help with recurring ASO work:

audit an app listing and prioritize fixes
improve titles, subtitles, keyword fields, descriptions, and release notes
research keywords and plan keyword coverage across locales
compare competitors and find gaps
prepare App Store featuring nominations or In-App Events
draft replies to reviews and analyze review sentiment
localize Figma screenshot JSON while preserving template tags
work with monetization, subscriptions, pricing, and availability drafts

Skills are most useful when the assistant can combine your product context, current ASO.dev data, and a clear workflow instead of answering from a generic prompt.

How It Works

Each skill is a folder with a SKILL.md file. After installation, your AI tool can discover those skills and use the right workflow for the task.

The recommended operating pattern is:

Ask the AI assistant for a specific ASO task.
Let the relevant skill gather context from the active ASO.dev project or files.
Review the proposed changes.
Ask the assistant to fill ASO.dev drafts only when you are ready.
Save or publish final changes from ASO.dev manually.

Install

Clone the repository first:

git clone https://github.com/asodevapp/skills.git
cd skills

Codex

Install the skills globally:

mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills"
cp -R skills/* "${CODEX_HOME:-$HOME/.codex}/skills/"

Restart Codex after copying the folders.

Claude Code

Use the Claude Code plugin marketplace flow:

/plugin marketplace add asodevapp/skills
/plugin install asodev-skills@asodev-skills

For a local copy instead:

mkdir -p "$HOME/.claude/skills"
cp -R skills/* "$HOME/.claude/skills/"

Restart Claude Code after installation.

Cursor

Install skills into the current project:

mkdir -p .cursor/skills
cp -R skills/* .cursor/skills/

Reload the Cursor window after copying.

Antigravity

Use the common global skills directory:

mkdir -p "$HOME/.gemini/antigravity/skills"
cp -R skills/* "$HOME/.gemini/antigravity/skills/"

Restart Antigravity or reload the workspace.

Connect to ASO.dev

The strongest workflow is Agent Skills plus the local AI Companion MCP in ASO.dev.

Open ASO.dev Settings -> AI Companion.
Start the local Companion server.
Use the ready config from ASO.dev or connect your AI tool to http://localhost:42111.
Ask the agent to read the current editor, review list, nomination draft, event draft, or project context.
Ask it to fill draft fields only after you approve the direction.

The local Companion usually runs on:

http://localhost:42111
http://127.0.0.1:42111

AI Companion MCP Setup Guide

ASO.dev API for Developers and Partners

Skill Areas

The skill pack covers several groups:

Tools and integrations: local MCP workflow, public API integration notes
Core workflow: ASO audits, current metadata fetch, validation, app context, analytics interpretation
Metadata and localization: Apple and Google metadata, keyword coverage, cross-localization, release notes, seasonal campaigns
Growth: competitor analysis, Apple Ads intelligence, featuring, In-App Events, CPP/PPO, screenshot briefs
Reviews: review sentiment, replies, complaint drafts, and rating strategy
Monetization: in-app purchases, subscriptions, pricing, availability, lifecycle and paywall strategy

Example Prompts

After installation, try prompts like:

Use ASO.dev skills to audit the current App Store metadata and give me the top 10 fixes.

Read the current ASO.dev editor through MCP and rewrite title, subtitle, and keywords for en-US.

Analyze unanswered low-star reviews and draft safe replies in the user's language.

Prepare an In-App Event for our next seasonal campaign and keep all metadata within App Store limits.

Localize this Figma screenshot JSON to all App Store Connect locales without changing template tags.

Verify

After installation, ask your AI tool:

What ASO.dev skills are available?

If the skills do not appear, restart the tool, reload the workspace, and check that each copied skill folder contains a SKILL.md file.