Docs Fields MCP Use cases TemplatesPricing Sign in Get API key
AI agents

Code and agent SDKs

Everything you need to call Isopleth from your own code or agent. Set two environment variables, pick your stack and paste. Every snippet reads the key from ISOPLETH_KEY; keep it on the server.

Environment
export ISOPLETH_KEY=paste_your_key_here
export ISOPLETH_BASE_URL=https://api.isopleth.dev

No key yet? Get a free API key (1,000 free lookups every month · no card).

Create or copy a key on your dashboard. Keys are shown once when you create them.

Drop-in snippets

REST, no SDK needed.

cURL
curl https://api.isopleth.dev/v1/fetch \
  -H "Authorization: Bearer $ISOPLETH_KEY" \
  -H "Content-Type: application/json" \
  -d '{"lat": 30.2672, "lng": -97.7431,
       "fields": ["nfhl_flood_zone", "elevation", "nearest_substation_distance_m"]}'

pip install requests

Python
import os, requests

r = requests.post(
    "https://api.isopleth.dev/v1/fetch",
    headers={"Authorization": f"Bearer {os.environ['ISOPLETH_KEY']}"},
    json={"lat": 30.2672, "lng": -97.7431,
          "fields": ["nfhl_flood_zone", "elevation", "nearest_substation_distance_m"]},
    timeout=60,
)
r.raise_for_status()
for name, f in r.json()["fields"].items():
    print(name, f["value"], f.get("unit"), f["source"])

Node 18+, Deno, Bun or a server route. Keep the key server side.

TypeScript
const r = await fetch("https://api.isopleth.dev/v1/fetch", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.ISOPLETH_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    lat: 30.2672, lng: -97.7431,
    fields: ["nfhl_flood_zone", "elevation", "nearest_substation_distance_m"],
  }),
});
const { fields } = await r.json();
for (const [name, f] of Object.entries(fields)) console.log(name, f.value, f.unit, f.source);

pip install openai. OpenAI calls the Isopleth MCP server for you.

OpenAI Responses
import os
from openai import OpenAI

client = OpenAI()
resp = client.responses.create(
    model="gpt-6-astra",
    tools=[{
        "type": "mcp",
        "server_label": "isopleth",
        "server_url": "https://api.isopleth.dev/mcp",
        "authorization": os.environ["ISOPLETH_KEY"],
        "require_approval": "never",
    }],
    input="What flood zone is 30.2672, -97.7431 in, and how far is the nearest substation?",
)
print(resp.output_text)

pip install openai-agents

OpenAI Agents SDK
import asyncio, os
from agents import Agent, Runner
from agents.mcp import MCPServerStreamableHttp

async def main():
    async with MCPServerStreamableHttp(
        name="isopleth",
        params={"url": "https://api.isopleth.dev/mcp",
                "headers": {"Authorization": f"Bearer {os.environ['ISOPLETH_KEY']}"},
                "timeout": 60},
        client_session_timeout_seconds=120,
    ) as isopleth:
        agent = Agent(name="Land analyst", instructions="Cite the source of every value.",
                      mcp_servers=[isopleth])
        result = await Runner.run(agent, "Land read for 38.71, -98.92: slope, soils, flood zone.")
        print(result.final_output)

asyncio.run(main())

pip install anthropic. Claude's MCP connector calls Isopleth for you.

Claude API
import os
import anthropic

client = anthropic.Anthropic()
msg = client.beta.messages.create(
    model="claude-sonnet-5-5",
    max_tokens=2000,
    betas=["mcp-client-2025-11-20"],
    mcp_servers=[{
        "type": "url",
        "name": "isopleth",
        "url": "https://api.isopleth.dev/mcp",
        "authorization_token": os.environ["ISOPLETH_KEY"],
    }],
    tools=[{"type": "mcp_toolset", "mcp_server_name": "isopleth"}],
    messages=[{"role": "user", "content": "Power read for 35.0844, -106.6504."}],
)
print(msg.content[-1].text)

pip install langchain langchain-mcp-adapters langchain-openai

LangChain / LangGraph
import asyncio, os
from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain.agents import create_agent

async def main():
    client = MultiServerMCPClient({
        "isopleth": {
            "transport": "streamable_http",
            "url": "https://api.isopleth.dev/mcp",
            "headers": {"Authorization": f"Bearer {os.environ['ISOPLETH_KEY']}"},
        }
    })
    tools = await client.get_tools()
    agent = create_agent("openai:gpt-6-astra", tools)
    out = await agent.ainvoke({"messages": [{"role": "user", "content": "Hazards read for 29.95, -90.07"}]})
    print(out["messages"][-1].content)

asyncio.run(main())

npm i ai @ai-sdk/mcp @ai-sdk/openai

Vercel AI SDK
import { createMCPClient } from "@ai-sdk/mcp";
import { generateText, stepCountIs } from "ai";
import { openai } from "@ai-sdk/openai";

const isopleth = await createMCPClient({
  transport: {
    type: "http",
    url: "https://api.isopleth.dev/mcp",
    headers: { Authorization: `Bearer ${process.env.ISOPLETH_KEY}` },
  },
});
try {
  const { text } = await generateText({
    model: openai("gpt-6-astra"),
    tools: await isopleth.tools(),
    stopWhen: stepCountIs(8),
    prompt: "Solar siting read for 35.08, -106.65. Cite sources.",
  });
  console.log(text);
} finally {
  await isopleth.close();
}

The hosted-tool snippets (OpenAI Responses, Claude API) let the model provider call https://api.isopleth.dev/mcp for you, so your code never handles tool calls. The SDK snippets (Agents SDK, LangChain, Vercel AI SDK) connect from your process and list Isopleth's 10 tools.

OpenAPI: GPT Actions, Postman and code generators

The full REST schema is at https://isopleth.dev/openapi.json (OpenAPI 3.1.0, CORS-enabled).

OpenAPI URL
https://isopleth.dev/openapi.json

For coding agents

/llms.txt is a short, plain-text map of the API, the MCP server and the docs, for Cursor, Claude Code, Codex and other agents that read it. Point your agent at it, or paste Read https://isopleth.dev/llms.txt and wire Isopleth into this project.

Prefer an app?

Claude, ChatGPT, Cursor, VS Code, Claude Code and Codex connect without code: see MCP server.

← PreviousMCP serverNext →Parcels