Small Business Intelligence MCP
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Small Business Intelligence MCPGive me a full teardown of Mom's Bakery in Denver"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Small Business Intelligence
by Brick & Mortar — a free, standalone remote MCP server for tearing down small businesses.
It ships analytical methodology, not data. Every tool returns a rigorously structured framework — a research procedure, an output schema, a quality rubric, caveats — and the calling model executes the research itself, with its own web search. This server never calls an external API, a database, or any data provider on your behalf. Your own AI subscription pays for all the inference; our marginal cost per call is effectively zero.
Live endpoint: https://sbi-mcp.small-business-intelligence-mcp.workers.dev/mcp
(no auth required — see Deploy for standing up your own copy).
The eight tools
Tool | What it does |
| Full structured teardown of one named business — digital presence, review signal, competitive position, pricing posture, visibility gaps, prioritized recommendations. The flagship tool. |
| Maps the local competitive set for a category + metro: true competitors vs. adjacent players, positioning matrix, saturation signals. |
| Mines public reviews: complaint taxonomy, theme extraction, sentiment trajectory, differentiators customers actually cite, red flags for buyers. |
| Local search presence audit: map-pack factors, listing consistency, category selection, site fundamentals — a scored checklist. |
| Builds a defensible local pricing comparison, including how to normalize across service bundles and what to do when prices aren't published. |
| Pre-diligence framework for brokers/buyers: SDE framing, multiple ranges (caveated, verify-live), red-flag checklist, seller questions. |
| Gap analysis for a category × metro: underserved demand, oversaturation, whitespace — from public signals only. |
| Assembles the outputs of prior tool calls into one polished, client-ready report, matched to the audience. |
Related MCP server: Scout Intel MCP
Why this architecture
See ARCHITECTURE.md for the full write-up. Short version:
@modelcontextprotocol/server
v2 (McpServer) + agents' stateless
createMcpHandler, mounted as a plain Cloudflare Worker fetch handler at
/mcp. No Durable Objects, no database, no external API calls — genuinely
zero marginal cost per call.
Local development
npm install
npm run typecheck # tsc --noEmit
npm run dev # wrangler dev — serves http://localhost:8787wrangler dev runs against Miniflare's local KV simulation, so the usage
ledger works out of the box with no Cloudflare account needed for local dev.
Testing with MCP Inspector
With wrangler dev running in a separate terminal:
# list all 8 tools
npx @modelcontextprotocol/inspector --cli --server-url http://localhost:8787/mcp \
--transport http --method tools/list --format json
# call one
npx @modelcontextprotocol/inspector --cli --server-url http://localhost:8787/mcp \
--transport http --method tools/call --tool-name business_teardown \
--tool-args-json '{"business_name":"Example Cafe","city_metro":"Saint Paul, MN"}' \
--format jsonOr drop --cli --format json for the interactive web UI (npm run inspector).
Confirm all 8 tools list with readOnlyHint: true and an outputSchema, and
that a tools/call against each returns structuredContent matching it —
verified this way during the build (no isError, no protocol-level
rejection, tools/list count = 8).
Manual smoke test (no Inspector needed)
curl -s -X POST http://localhost:8787/mcp \
-H "content-type: application/json" \
-H "accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"business_teardown","arguments":{"business_name":"Example Cafe","city_metro":"Saint Paul, MN"}}}'Deploy
wrangler login # or set CLOUDFLARE_API_TOKEN
wrangler kv namespace create USAGE_LEDGER # once — copy the id into wrangler.jsonc
npm run deployShips to the free *.workers.dev subdomain by default — no DNS work, no
paid plan. A custom domain is a one-line routes addition in
wrangler.jsonc later.
Usage stats
Every real tool call increments an aggregate, non-identifying counter
(stats:{tool_name}:{date} in the USAGE_LEDGER KV namespace — see
src/middleware/stats.ts, disclosed in /privacy and /docs). To see
which tools actually get used:
export $(grep -v ^# .env | xargs)
python3 scripts/usage_stats.py # last 30 days
python3 scripts/usage_stats.py --days 7 # last 7 daysMonetization (built, not active)
Every tool passes through a single middleware seam
(src/middleware/context.ts). Today, POLICY_MODE=allow_all (the shipped
default in wrangler.jsonc) never denies a call. Flipping it to metered
activates a per-identity daily call limit (POLICY_METERED_DAILY_LIMIT),
backed by a Workers KV usage ledger, with clean, plain-language denials
(never a broken-looking protocol error) that include an upgrade link. No
authentication is required at launch; the seam is architected so OAuth 2.1 +
PKCE can be added later without restructuring — see src/oauth/stub.ts and
ARCHITECTURE.md's "OAuth: stubbed, not mounted" section.
Repo structure
src/
├── index.ts # Worker entry — routes / , /privacy, /mcp
├── server.ts # createServer(): builds McpServer, registers all 8 tools
├── env.ts # Env (Worker bindings/vars) type
├── middleware/ # identity resolution, KV usage ledger, policy, withPolicy seam
├── oauth/ # OAuth 2.1 discovery handlers — written, not mounted (see above)
├── tools/ # one file per tool, all sharing the FrameworkPayload shape in types.ts
└── pages/ # landing (/) and privacy (/privacy) page HTMLLicense
MIT — see LICENSE.
Privacy
See PRIVACY.md (or /privacy on the live deploy).
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-qualityCmaintenance250+ AI-powered MCP tools: research, write, code, translate, scrape, sentiment, vision, RAG, agent memory, marketplace, trading signals, and more. 15 models across 7 providers. Pay-per-use via API key or x402 USDC micropayments.Last updated2MIT
- AlicenseAqualityCmaintenanceWeb intelligence MCP server for AI agents. 7 tools for SERP analysis, competitor research, market trends, content gap analysis, keyword insights, audience discovery, and citation tracking.Last updated72AGPL 3.0
- AlicenseAqualityAmaintenanceHosted Amazon market-intelligence MCP for Claude and ChatGPT: query brands, sellers, ASINs, under-competed niches, the cross-seller operator network, observed buy-box history, and Amazon/Walmart cross-marketplace overlap. 65 read-only research tools over a pre-collected research dataset.Last updated72MIT
- Alicense-qualityBmaintenanceMake AIs recommend your business. Measures how AI assistants answer the questions your market asks, which sources they cite, and gives the levers to shape those answers. 43 tools on a hosted Streamable HTTP endpoint at https://mcp.epovest.com/mcpLast updatedMIT
Related MCP Connectors
100+ MCP tools for AI agents: content metadata, trade intelligence, business-expertise analysis.
Social media analytics, video analysis, and competitor intel for any MCP-compatible AI agent.
Hosted MCP with 91 agent tools: X, domains, SEO, Maps, Trends, Search, YouTube, TikTok, and more.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/2016judea/small-business-intelligence-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server