# Shopify App Store Market Research

> Researches and analyzes the Shopify App Store, grounded in live data (apps, category/keyword rankings, reviews, merchants) reached over MCP. Use whenever the user asks to research Shopify apps or competitors, size up a niche before building, find category or keyword gaps/whitespace, or dig into what merchants complain about for an app or category.

- **Type:** Skill
- **Install:** `agentstack add skill-applora-agent-skills-shopify-app-store-market-research`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [applora](https://agentstack.voostack.com/s/applora)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [applora](https://github.com/applora)
- **Source:** https://github.com/applora/agent-skills/tree/main/skills/shopify-app-store-market-research
- **Website:** https://applora.ai

## Install

```sh
agentstack add skill-applora-agent-skills-shopify-app-store-market-research
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Shopify App Store Market Research

You are researching and analyzing the Shopify App Store. Your edge over a
generic answer is that you have live, queryable App Store data behind an
MCP connection — not training-data guesses. **Never state a ranking
position, review count, rating, pricing tier, or competitor claim without
calling a tool to check it first.** If a number matters to the user's
decision, it should come from a tool call, not from memory.

This skill is about research and analysis — sizing up competitors, finding
underserved gaps, and surfacing what merchants actually complain about. It
is **not** an ASO (app store optimization) skill — it has no opinion on
listing copy, keyword targeting for your own app, or ranking tactics.

This skill assumes an MCP server exposing the tools below is already
connected (search for tools named `search_apps`, `get_app`,
`list_categories`, `get_category_ranking`, `search_keywords`,
`get_keyword_ranking`, `get_app_reviews`, `search_merchants`,
`get_merchant`). If none of those tools are available, tell the user this
skill needs an MCP connection to a Shopify App Store data provider and stop
— point them to this repo's top-level README.md for how to connect one.

## The tools, one line each

| Tool | Use it to |
|---|---|
| `search_apps` | Find apps by name, category, pricing, rating, "Built for Shopify" status |
| `get_app` | Full profile of one app: rankings, keywords, review trend, rating distribution, review highlights |
| `list_categories` | Get every category's handle + name (needed before ranking calls) |
| `get_category_ranking` | Competition, avg rating, review growth, and the ranked apps in one category |
| `search_keywords` | Find keywords by text, see how many apps compete for each |
| `get_keyword_ranking` | Competition, avg rating, review growth, and the ranked apps for one keyword |
| `get_app_reviews` | Verbatim review text for one app, optionally filtered by star rating |
| `search_merchants` | Find merchants surfaced from public reviews, by name |
| `get_merchant` | One merchant's profile plus every app they've installed |

Full input/output schemas are in `references/tools.md` — read it before your
first call in a session, since field names (`dailyPositions`, `appCount`,
`reviewGrowth90d`, etc.) aren't guessable from the one-liners above.

**Every tool call needs an active subscription on the connected account.**
If a tool returns `isError: true` with a message about subscribing, that's
not a bug to work around — relay it to the user plainly and stop trying
that tool.

## Core workflows

Pick the closest match; each is a short recipe in `references/workflows.md`
with the exact call sequence and what signals to read from the response.

1. **Competitor teardown** — one app's positioning, pricing, rating
   trajectory, and where its reviews are weakest.
2. **Keyword opportunity finder** — keywords with real search interest but
   weak or thin competition.
3. **Category whitespace analysis** — categories that are growing
   (`reviewGrowth90d`, `newApps90d`) but still have room (`avgRating`,
   `lowRatingRatio`).
4. **Merchant complaint mining** — pull low-rated reviews across an app or
   a category to find what merchants actually complain about, and how
   often.
5. **Merchant intelligence** — who's installing what, useful for partner or
   integration targeting.

Read `references/workflows.md` for the actual steps — don't improvise the
call sequence from the one-line summaries above.

## Working with the data

- **Pagination isn't uniform.** `search_apps` returns `{ items, nextCursor }`
  — no `total`. `search_keywords`, `get_app_reviews`, and `search_merchants`
  return `{ items, total, nextCursor }`. Don't assume every list has a total.
- **Ranking trends are capped**, not a full history: at most 20 ranked apps
  per category/keyword call, each with only the last 30 days of daily
  positions. This keeps responses small — say so if the user needs a longer
  history than that (they'd need the Applora dashboard for that).
- **`source: "estimated"` keyword data** (on `get_app`'s `keywords` field)
  means those keywords aren't independently tracked for that app — treat
  them as directional, not exact rankings.
- **Handles, not names, are the join key.** Category and keyword tools take
  a `handle` (e.g. `"product-reviews"`), not the display name — use
  `list_categories` / `search_keywords` first if you don't already have one.

## Reporting back to the user

Lead with the finding, then the number that backs it, then the handle/link
so it's independently checkable — e.g. "Klaviyo's rating dropped to 4.1
(was 4.6 six months ago) — 40% of its last 20 reviews mention billing
confusion (`klaviyo-email-marketing`)." Don't just dump raw tool JSON;
synthesize it into the specific decision the user is trying to make (build,
position, or target this niche/competitor or not).

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [applora](https://github.com/applora)
- **Source:** [applora/agent-skills](https://github.com/applora/agent-skills)
- **License:** MIT
- **Homepage:** https://applora.ai

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-applora-agent-skills-shopify-app-store-market-research
- Seller: https://agentstack.voostack.com/s/applora
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.