analyze_competitorsLABSCO SUMMARY
The four questions you ask about a business you do not know yet.
Competitors, web presence, reviews and market size, each one call instead of an afternoon of tabs. The analysis is model-generated, which is why the server you run asks for your own OpenAI key.
What it is
Four business-research tools over MCP: competitive analysis, website presence scoring, review analysis and market research. You run it yourself on Bun.
What you get
- A competitive landscape with market positioning, SWOT insights and strategic recommendations (analyze_competitors)
- A website scored 0 to 100 across SEO, performance, content, social media and trust signals (score_web_presence)
- Online reviews aggregated with sentiment analysis, theme extraction and reputation insights (analyze_reviews)
- Industry research with market sizing, trends, opportunities and customer segment analysis (market_research)
Requirements
Bun, your own OpenAI API key and an admin secret you choose, both set in .env. The server mints its own API keys: call /api/keys/provision with the admin secret, then send the key as an x-api-key header to http://localhost:4200/mcp.
Setup effort
Build from source — git clone https://github.com/ezbiz-services/mcp-business-intelligence.git
USE CASES
01
Sizing up who else is doing this
Name the business and the location and get the competitive landscape back with positioning and SWOT, which is the research pass you would otherwise do by reading twenty homepages.
02
Telling a client what is wrong with their site
The score breaks into SEO, performance, content, social and trust signals, so the conversation is about which of the five is dragging rather than about a single number.
03
Finding out what customers actually complain about
Reviews are aggregated and themed rather than counted, so what comes back is the recurring problem behind the rating instead of an average.
04
Checking whether a market is worth entering
Market sizing, trends, opportunities and customer segments in one call, which is enough to decide whether the idea deserves a real research budget.
4 tools
Listed in project docs. Your client may show a different list.
score_web_presenceanalyze_reviewsmarket_researchBEFORE YOU INSTALL
What to know first
Tools you get
4 tools named in the project's documentation.
Requires
- Bun
- An OpenAI API key
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
This one is built from source rather than installed with a single command.
Server setup
TERMINAL
git clone https://github.com/ezbiz-services/mcp-business-intelligence.git
cd mcp-business-intelligence
bun install
cp .env.example .envSet OPENAI_API_KEY and ADMIN_SECRET in .env
TERMINAL
bun run server.tsLeave the server running. In a second terminal, mint an API key with your ADMIN_SECRET — the key in the response goes in the x-api-key header of the client entry below:
TERMINAL
curl -X POST http://localhost:4200/api/keys/provision -H "x-admin-secret: <YOUR_ADMIN_SECRET>" -H "Content-Type: application/json" -d '{"email":"<YOUR_EMAIL>","tier":"business"}'MCP Clients
Cursor
CURSOR · JSON
{
"mcpServers": {
"bizintel": {
"url": "http://localhost:4200/mcp",
"headers": {
"x-api-key": "<YOUR_API_KEY>"
}
}
}
}
