crewai-sibfly

A CrewAI tool for SibFly — measured ground motion (subsidence & uplift) for any US address, in mm/year, from NASA OPERA Sentinel-1 satellite radar (InSAR). Measured, not modeled.

Negative mm/year = the ground is sinking. Give your agents ground truth about whether the land under an address is sinking, rising, or stable.

Install

pip install -U crewai-sibfly
export SIBFLY_API_KEY="sf_live_..."   # free key at https://sibfly.com (agents can self-register)

Use

from crewai import Agent
from crewai_sibfly import SibflyGroundMotionTool

analyst = Agent(
    role="Property risk analyst",
    goal="Tell the user if the ground under an address is sinking",
    tools=[SibflyGroundMotionTool()],
)

Or call it directly:

from crewai_sibfly import SibflyGroundMotionTool

tool = SibflyGroundMotionTool()                       # reads SIBFLY_API_KEY
tool.run(address="1100 Congress Ave, Austin, TX")
tool.run(lat=36.098, lon=-119.56, dry_run=True)       # FREE coverage + price check

Pricing (agent-friendly)

• $0.40 per covered report. Out-of-coverage, no-data, too-stale, and low-confidence results are free (cost_usd: 0).
• Pass dry_run=True for a free coverage + price check before spending.
• Route logic on the stable assessment_code (rapid_subsidence / notable_subsidence / stable / mild_uplift / strong_uplift), not the human-readable string.

Also available

LangChain (langchain-sibfly), LlamaIndex (llama-index-tools-sibfly), n8n (n8n-nodes-sibfly), a hosted MCP server at https://sibfly.com/mcp, and a plain REST API. See https://sibfly.com/docs.

Data: NASA OPERA DISP-S1 (contains modified Copernicus Sentinel data). Screening estimate from satellite radar, not an engineering survey. MIT-licensed integration.