mcpField
W
Whereabot

Whereabot

Remote
whereabot.ioMaps & Travel0Updated 1d ago
RepositoryVisit website

About Whereabot

Whereabot is a location-request relay for AI agents, paired with a phone app that returns a current GPS fix after consent. It is intended for people who want agent access to their location without retaining location history, and can be self-hosted.

Key features

  1. 01On-demand phone location requests
  2. 02Consent-based location responses
  3. 03No location-history retention
  4. 04Rust relay with HTTP and MCP
  5. 05Tauri app for iOS and Android
  6. 06Self-hostable relay and Supabase

Use cases

  1. 01Let an agent ask for a current location
  2. 02Share location with another person's agent
  3. 03Run a private location relay

MCP support

The relay exposes location requests to MCP clients over HTTP, alongside an HTTP interface. The phone connects to the relay over WebSocket and responds with a current GPS fix.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://relay.whereabot.io/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

Whereabot

Let your AI agent ask your phone where you are, on demand. You host the relay. Sign-in and user data live in Supabase.

flowchart LR
  Agent["AI agent"] -->|"MCP or HTTP"| Relay["Relay (Rust)"]
  Relay -->|WebSocket| Phone["Phone app (Tauri)"]
  Phone -->|"GPS fix"| Relay
  Relay -->|"answer"| Agent

The relay keeps no location history. It sends a ping, the phone reads its GPS now, and the answer goes back to the agent. What is stored, for how long, and who sees your location: privacy.

Status

Early MVP, in beta. iOS answers in the background through push. Android answers only while the app is open. See the roadmap.

Repo layout

PathWhat
server/Rust relay (axum). Agents call it over HTTP or MCP; the phone holds a WebSocket to it.
app/Tauri v2 phone app (Android / iOS). Signs in, then answers pings with a GPS fix.
supabase/Database migrations and tests.
infra/OpenTofu for the dev and prod environments.
scripts/Local dev, end-to-end check, deploy, hook install.
docs/Source of the docs site.
site/Landing page and OAuth consent page.
docs-site/VitePress build of the docs site.
design-system/, brand/Shared CSS tokens and components; logo files and usage rules.

Connect your agent

  1. Install the app: see Get started on whereabot.io, or install from the browser.

  2. Sign in with Google on the phone.

  3. Add the relay to your agent:

    claude mcp add --transport http whereabot https://relay.whereabot.io/mcp

  4. In Claude Code run /mcp and sign in with the same Google account as on the phone. You never paste a token.

  5. On Android, keep the app open while the agent asks, until Android push ships.

Relay URLs: live URLs. The tool and its arguments: protocol.

Self-host

Run your own relay and Supabase project with the OpenTofu code in infra/. Steps: infra/README.md. Build the app with your Supabase and relay values (app/.env.example), and add your relay to your agent. To run everything on one computer first: local quick start.

Docs

All docs, with the generated MCP and wire references: whereabot.io/docs. In this repo:

Community

License

MIT. Third-party material: THIRD_PARTY_NOTICES.md.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.