Clean up a messy address.

Free-text in, structured out. Fixes typos, expands abbreviations, infers missing fields, and scores confidence.

mcp.footstep.ai / parse_address

Why you need it

Why your agent needs this.

Radio transcripts, voice notes, and forwarded messages don't arrive as structured addresses. 'last seen near the Crown pub on Bridge St' is what the agent gets. parse_address turns that into fields a geocoder can use, with corrections itemised so the agent can flag uncertainty back to the operator.

What it does for you

What your agent gets back.

Structured output the agent can hand directly to geocode or batch_geocode.
Itemised corrections list ('st → Street', 'ave → Avenue') the agent can disclose.
Per-row confidence score so the agent routes low-confidence parses to review.
Composes with batch_geocode for bulk pipelines and reverse_geocode for verification.
LLM-grounded: handles typos, missing fields, and abbreviations that a regex parser misses.

Built for

For SAR teams For public-safety GIS For AI developers For integrators

Try it

Plug it into your agent.

Add Footstep to your MCP-aware runtime. Your model picks the tool at runtime; you wrote none of the orchestration.

Read the full reference

Prompt that triggers parse_address

"Standardise this CSV of addresses with typos and abbreviations."

claude_desktop_config.json

{
  "mcpServers": {
    "footstep": {
      "url": "https://mcp.footstep.ai",
      "headers": {
        "x-api-key": "sk_live_your_key_here"
      }
    }
  }
}

Related capabilities.

geocode

Turn an address, landmark, or place name into coordinates. Returns ranked candidates with confidence scores.

batch_geocode

Run up to 1,000 addresses through geocoding in a single call. Returns coordinates and confidence for each.

reverse_geocode

Given a coordinate, return the address or place label. Useful for labelling pins and incident reports.

Ready to call it?

£5 free credit on signup. One auth header, every major runtime.

Get an API key See the REST endpoint