Getting Started

Overview #

GapQuery scans 11 app ecosystems and surfaces the opportunities hiding in them: underserved categories, poorly-rated incumbents, overpriced tools, and missing integrations. It's for developers and micro SaaS founders who want real data behind their next idea.

You interact with the data through an MCP server that plugs into Claude Code. A companion research skill walks Claude through a structured discovery and analysis workflow so you get actionable results, not just raw numbers.

Prerequisites #

  • A GapQuery account with an active purchase
  • Claude Code installed and configured
  • An API key (create one under Settings > API Keys after purchase)

Get Your API Key #

After purchasing, go to Settings > API Keys and create a new key. The ecosystems you can access depend on your purchase tier:

Tier Price Ecosystems
Bundle $99 one-time 5 ecosystems of your choice
Unlimited $199 one-time All 11 ecosystems
Monthly $49/mo All 11 ecosystems (converts to lifetime after 4 payments)

Select your ecosystems after purchase in Settings > API Keys. You can change your selection at any time.

Connect Claude Code #

The GapQuery MCP server gives Claude Code direct access to ecosystem data and gap analysis tools. Add this to your project's .mcp.json file:

json
{
  "mcpServers": {
    "gapquery": {
      "type": "http",
      "url": "https://www.gapquery.com/mcp/gapquery",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Replace YOUR_API_KEY with the key you created in the previous step.

The MCP server uses the streamable HTTP transport. Make sure your version of Claude Code supports MCP.

Verify Connection #

Open Claude Code in your project and ask it to list available ecosystems. If it comes back with data, you're connected.

text
You: "What ecosystems are available in GapQuery?"
Claude: Lists ecosystems with slugs and app counts
If Claude can't find the tools, check that .mcp.json is in your project root and restart Claude Code.

Install the Research Skill #

The gapquery-research skill gives Claude Code a structured workflow for discovering and researching opportunities. It builds on top of the MCP tools.

Install the skill with one command:

bash
npx skills add northify/skills --skill gapquery-research
The skill is just a markdown file that Claude Code reads. There's nothing else to install.

Updating the skill #

We ship skill improvements regularly. When the GapQuery MCP server returns a newer version than your local skill, Claude will tell you. At that point, re-run the same install command:

bash
npx skills add northify/skills -y
Use add, not update. The npx skills update command only works for skills installed by recent versions of the skills CLI; npx skills add northify/skills -y always works regardless of when you first installed.

Once installed, just ask Claude to find opportunities or research an ecosystem. The skill activates automatically and walks Claude through a 4-phase discovery workflow, then a 6-area deep research workflow.

What the Skill Does #

Discovery (Phases 1 through 4)

  1. Ecosystem Scan: Broad analysis across all dimensions
  2. Creative Synthesis: Connects raw signals into opportunity ideas
  3. Drill-down: Zooms into the most promising signals
  4. Present & Select: Ranked list of opportunities for you to pick from

Research (6 Areas)

  1. Market Validation: TAM/SAM, growth trends, timing
  2. Competition: Direct and indirect competitors, your advantages
  3. Revenue: Pricing model, estimated MRR, key assumptions
  4. Technical: APIs, complexity, dev time, stack
  5. Keywords: SEO difficulty, content strategy
  6. Go-to-Market: Launch strategy, channels, partnerships

Next Steps #

You're all set. Here's the fastest path to your first opportunity:

  1. Pick an ecosystem you know well
  2. Ask Claude to scan it for opportunities
  3. Save the top opportunities to your pipeline
  4. Run deep research on your shortlist
  5. Build what the data says is missing
Read the Discovery Guide to learn how GapQuery's 8 opportunity dimensions work, or jump to the Research Guide for details on the 6-area deep research workflow.