MCP.Market

Permits MCP server

by addressintel.co·co.addressintel/permits·v1.1.2

SF Peninsula building permits and parcel buildability: SB 9, ADU and redevelopment leads.

B77/100grade B
What users say
No reviews yet
Be the first
Safety scan
B77/100

full report

Adoption
Growing

0 stars77 downloads/wk

Reviews

Write one

Nobody has reviewed Permits yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Permits tools (6)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • get_property_intelligence

    Get detailed AI intelligence for a specific property. Takes an id from list_market_signals. Returns teardown and flippability scores, a market heat index, an AI briefing, comparable-sale counts with an average price per sqft, and parcel enrichment (zoning, lot size, development class). Those scores rank candidates; none of them is a return. This endpoint carries no ROI figure at all: for that, rea

  • list_adu_leads

    List parcels ranked by ADU (accessory dwelling unit) feasibility, including estimated max units, buildable square footage and the lot geometry behind the score.

  • list_market_signals

    Search for high-signal real estate investment opportunities (teardowns, flips). Use this to find properties before diving into specific details. Each result carries proformaROI, a net teardown ROI computed by arithmetic rather than by a model, with a proforma object holding its inputs and assumptions ($400/sqft build, 5% cost of sale, build sized to the zoning envelope). Quote those assumptions al

  • list_redevelopment_leads

    List underbuilt parcels: small, older homes using a low share of their allowable building envelope on sizable lots. Ranked by redevelopment score, with the FAR headroom that produced it.

  • list_sb9_inventory

    List active listings that are eligible for an SB 9 lot split or duplex conversion, with the per-parcel block reasons where they are not. California-only; this is a parcel-level buildability read, not a generic lead list.

  • search_permits

    Search issued building permits across the SF Peninsula and adjacent Silicon Valley cities. Filter by city, keyword (address, project type, scope description, contractor, architect or permit number) and minimum valuation. Use this to find what construction has actually been permitted at an address or in a city.

Public scan report

scanner v0.1.10 · 2026-11-02 · same rubric, same numbers if you re-run it

no findings
  • Code scan3 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitystatic API keys via environment variables6/15
  • Maintenancelast push 75 days ago12/15
  • Maintainer identitynamespace and repository owner differ; GitHub account older than a year; website matches verified namespace7/10
Overall 77/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Permits compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
77/100
higher than 50% of 48,255 servers
Tools
6
the typical server has 7
Tools that change things
0%
typical: 0%. Fewer write tools means less an agent can break
Downloads a week
77
more than 56% of 12,191 servers with a package

What its tools let an agent do: 6 read

Install Permits in Claude Code, Cursor or VS Code

Runs npx -y addressintel-mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add permits -- npx -y addressintel-mcp
Add to Cursor

What the publisher says

From the Permits repository's README, as published. We do not edit it. Read it on GitHub

AddressIntel MCP server

A read-only Model Context Protocol server that puts SF Peninsula building-permit and parcel-buildability data inside Claude, ChatGPT, or any other MCP-capable agent.

It is a thin client over the public AddressIntel API, so it inherits that API's auth, rate limits and tier ceilings. Every tool is a GET. Nothing here can write.

Tools

The permit and parcel tools are the differentiated half: permit-level detail and parcel-level buildability for the Peninsula, rather than a nationwide owner list.

Install

No install step: npx fetches it on demand.

Claude Code

claude mcp add addressintel \
  -e ADDRESSINTEL_API_KEY=your_key \
  -- npx -y addressintel-mcp

Claude Desktop

Add to claudedesktopconfig.json:

{
  "mcpServers": {
    "addressintel": {
      "command": "npx",
      "args": ["-y", "addressintel-mcp"],
      "env": {
        "ADDRESSINTEL_API_KEY": "your_key"
      }
    }
  }
}

From source

For working on the server itself:

git clone https://github.com/RantumBits/addressintel-mcp.git
cd addressintel-mcp
npm install
npm run build     # emits dist/index.js

Then point the client at node /absolute/path/to/addressintel-mcp/dist/index.js instead of the npx command above.

Restart the client, then ask it something like "what demolition permits were issued in Menlo Park this year?" or "which Palo Alto listings are SB 9 eligible?"

Configuration

Development

npm run dev       # run from source over stdio
node test_mcp.mjs # smoke test: connect, list tools, call one

Tools are declared in one table in src/index.ts. Each entry carries its zod schema and its advertised JSON Schema side by side, so adding a tool means adding one entry, and the two shapes cannot drift apart.

Nothing above is checked by us. What we check is on the safety report.

Permits: common questions

Is Permits MCP server safe?
Mostly: it is graded B (77/100). Read the Permits safety report
How do I install Permits?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Permits need an API key?
Yes. The registry entry asks for ADDRESSINTEL_API_KEY.
Is Permits maintained?
The last commit was 77 days ago (2026-08-19). The latest release is v1.1.2.
How does Permits compare with other MCP servers?
safety score 77/100 (higher than 50% of 48,255 servers); tools 6 (the typical server has 7); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.