Skip to main content
Glama
colinterface

Bay Wheels MCP Server

by colinterface
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Bay Wheels MCP Server

This is an MCP server that provides access to Bay Wheels realtime bikeshare data.

Features

  • Find nearest bike (standard or ebike)

  • Find nearest dock with available spaces

  • Supports checking for free bikes (dockless) when looking for a single bike

Related MCP server: mcp-citybikes

Setup

For Claude Desktop (Local Development)

Add this to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "bay-wheels": {
      "command": "/opt/homebrew/bin/uv",
      "args": [
        "--directory",
        "/path/to/bay-wheels-mcp",
        "run",
        "server.py"
      ]
    }
  }
}

Make sure to update the path to match your local installation directory.

Manual Testing (stdio)

You can run the server directly for testing with Claude Desktop:

uv run server.py

Deployment

Docker Deployment

Quick Start

# Build the image
docker build -t bay-wheels-mcp .

# Run the container
docker run -p 8000:8000 bay-wheels-mcp

# Test the health check
curl http://localhost:8000/health

Using Docker Compose

# Start the server
docker-compose up -d

# Check logs
docker-compose logs -f

# Stop the server
docker-compose down

Environment Variables

  • PORT - Server port (default: 8000)

  • HOST - Bind host (default: 0.0.0.0)

Health Check

The server exposes a health check endpoint at /health for container orchestration:

curl http://localhost:8000/health
# Response: {"status":"healthy","service":"bay-wheels-mcp","version":"0.1.0"}

Platform-Specific Deployment

AWS ECS/Fargate

  1. Push image to ECR:

docker build -t bay-wheels-mcp .
docker tag bay-wheels-mcp:latest <aws-account>.dkr.ecr.<region>.amazonaws.com/bay-wheels-mcp:latest
docker push <aws-account>.dkr.ecr.<region>.amazonaws.com/bay-wheels-mcp:latest

  1. Create ECS task definition with health check enabled

  2. Deploy as ECS service with load balancer

Google Cloud Run

gcloud builds submit --tag gcr.io/<project-id>/bay-wheels-mcp
gcloud run deploy bay-wheels-mcp --image gcr.io/<project-id>/bay-wheels-mcp --port 8000

Fly.io

fly launch --dockerfile Dockerfile
fly deploy

Azure Container Instances

Use Azure Portal or Azure CLI to deploy the Docker image with port 8000 exposed.

Connecting Mobile Apps

The deployed server uses StreamableHTTP transport. Configure your MCP client to connect to:

URL: https://your-deployed-server.com/mcp

Note: The MCP endpoint is at /mcp, not the root path.

See MCP documentation for client integration details.

Testing

Testing the Deployed Server

The simplest way to test is using the health check endpoint:

# Test that the server is running
curl https://your-server.com/health
# Should return: {"status":"healthy","service":"bay-wheels-mcp","version":"0.1.0"}

For full MCP protocol testing, use an MCP client (Claude Desktop, mobile app, or custom client). The StreamableHTTP transport requires session management and proper header negotiation which is best handled by official MCP clients.

Testing with MCP Clients

The best way to test the deployed server is to configure it in your MCP client:

Claude Desktop (Remote Server)

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "bay-wheels-remote": {
      "url": "https://your-server.com/mcp",
      "transport": "streamable-http"
    }
  }
}

Mobile App

Configure your mobile app's MCP client to connect to:

https://your-server.com/mcp

Then test the tools by asking Claude:

  • "Find me the nearest Bay Wheels bike near the Ferry Building in SF"

  • "Where can I return a bike near Dolores Park?"

Tools

find_nearest_bike

Finds the nearest bike availability.

  • latitude: float

  • longitude: float

  • count: int (default 1)

  • bike_type: str (optional, "classic_bike" or "electric_bike")

find_nearest_dock_spaces

Finds the nearest dock with return spaces.

  • latitude: float

  • longitude: float

  • count: int (default 1)

Available Tools

2 tools
find_nearest_bikeA

Find the nearest dock with at least N available bikes, or free bike locations if N=1.

Args: latitude: The latitude of the search location. longitude: The longitude of the search location. count: The number of bikes needed (default 1). bike_type: Optional type of bike ('electric_bike' or 'classic_bike'). If None, any type.

ParametersJSON Schema
NameRequiredDescriptionDefault
countNo
latitudeYes
bike_typeNo
longitudeYes

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It discloses the core behavioral logic: requiring at least N available bikes, special handling for N=1 ('free bike locations'), and optional bike type filtering. It goes beyond a simple verb-noun description, though it does not mention potential failure modes, sorting details, or what 'free bike locations' exactly entails. Given the read-only nature implied by 'Find', this is reasonably transparent.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is highly concise: a two-sentence summary followed by a structured parameter list. Every sentence adds value, and the front-loaded core behavior is immediately understandable. No unnecessary fluff.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has an output schema, the description does not need to explain return values. It explains the main behavior and all parameters, though it does not address edge cases like no available docks. The sibling tool is not mentioned, which slightly reduces completeness, but the core context is adequate for an agent to invoke the tool correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema has 0% description coverage, so the description must compensate. It provides an Args block that explains every parameter: latitude, longitude, count (with default), and bike_type (listing acceptable values and meaning of None). This fully explains the parameters beyond the schema's basic type definitions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states a specific verb ('Find') and resource ('nearest dock with at least N available bikes'), including a special case for N=1. It effectively distinguishes from the sibling tool find_nearest_dock_spaces by focusing on bike availability rather than dock spaces.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies the tool is for finding bikes by location and count, but it does not explicitly state when to use this tool versus the sibling find_nearest_dock_spaces, nor does it provide exclusions or alternative recommendations. The usage context is clear from the description but not explicitly contrasted.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

find_nearest_dock_spacesA

Find the nearest dock with at least N available return spaces.

Args: latitude: The latitude of the search location. longitude: The longitude of the search location. count: The number of spaces needed (default 1).

ParametersJSON Schema
NameRequiredDescriptionDefault
countNo
latitudeYes
longitudeYes

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A3.5/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It does not disclose whether the tool is read-only, whether it returns multiple docks or just the nearest, what happens if no dock has the required spaces, or if results are sorted by distance. The description leaves key behavioral aspects open to interpretation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise and front-loaded with the core purpose. The parameter list is structured and avoids fluff, though it partly duplicates the schema, it earns its place by compensating for the schema's lack of descriptions.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple 3-parameter finder with an output schema, the description is mostly complete but lacks usage guidance and behavioral edge cases. The presence of an output schema partially satisfies return value disclosure, but the absence of annotations leaves gaps in understanding tool behavior.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With schema coverage at 0%, the description must explain parameters. It adds minimal value by restating latitude/longitude as 'search location' and specifying count as 'number of spaces needed (default 1)'. The count explanation adds meaning, but latitude/longitude descriptions are largely tautological.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states what the tool does: 'Find the nearest dock with at least N available return spaces.' This is a specific verb+resource+constraint and distinguishes from the sibling tool find_nearest_bike.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage for finding docks with return space availability but does not explicitly state when to prefer this over find_nearest_bike or provide any exclusions. No alternatives or when-not guidance is given.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 2 tool updatesv0.1.0
    • First observedfind_nearest_bike
    • First observedfind_nearest_dock_spaces
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.9/5.0

Scored across 2 tools

Disambiguation5/5

The two tools have clearly distinct purposes: one locates available bikes for pickup, the other locates available return spaces. There is no overlap or ambiguity between them.

Naming Consistency5/5

Both tool names follow the consistent pattern 'find_nearest_' followed by the object ('bike' or 'dock_spaces'). This creates a predictable and coherent naming convention.

Tool Count3/5

With only two tools, the server feels thin but is appropriately scoped for a simple find-nearest utility. It is borderline, as it might benefit from additional tools like station list or trip planning, but it is not excessive.

Completeness4/5

The pair covers the two core needs of a bike-sharing user: finding bikes to rent and finding spaces to return. Minor gaps exist (e.g., station details or real-time status), but for the apparent minimal scope, it is mostly complete.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • Citi Bike MCP Server

    sjhangiani12
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables users to find the nearest Citi Bike stations with real-time availability data for electric and classic bikes. Provides distance-sorted results with detailed station information including bike availability, dock capacity, and operational status.
    -

  • Bixi MCP Server

    sevesogit
    A
    license
    Not graded
    quality
    D
    maintenance
    Provides real-time access to Montreal's Bixi bike-sharing data through the GBFS API. Enables querying station information, availability, nearby stations, and system alerts.
    MIT

  • mcp-bart

    pipeworx-io
    A
    license
    Not graded
    quality
    B
    maintenance
    Provides real-time San Francisco Bay Area Rapid Transit data, enabling queries about BART schedules, routes, and station information through natural language.
    4 npm
    MIT