Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@DiSH MCP Server Find a room for a 1-hour meeting tomorrow at 2 PM."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server for the DiSH room booking site. It exposes tools to check availability, book rooms, and cancel bookings so assistants like Cursor or Claude can manage your reservations.
Search room availability across your DiSH locations
Create bookings on your behalf
Cancel or reschedule existing bookings
Pair with calendar tools (e.g., Google Calendar MCP) to coordinate bookings with your calendar availability.
Example prompts:
"Book meeting rooms for all my standups this week."
"Reschedule my 1-1 with John to tomorrow afternoon when we’re both free and a room is open."
"Book a meeting room for all my customer demos for the next 2 months."
Related MCP server: nexudus-mcp
Python 3.10+
uv for dependency management
uv syncRequired:
TEAM_ID — your DiSH team ID
MEMBER_ID — your DiSH member ID
Authentication:
DISH_COOKIE — your DiSH connect.sid session cookie
DISH_EMAIL — optional, enables non-interactive login when the cookie is missing or expired
DISH_PASSWORD — optional, enables non-interactive login when the cookie is missing or expired
Optional transport configuration (for HTTP mode):
MCP_TRANSPORT — stdio (default) or http
MCP_PORT — port for HTTP transport (default: 8000)
MCP_HOST — host for HTTP transport (default: 127.0.0.1)
If you want Claude to keep working when connect.sid expires, put these values in .env:
TEAM_ID=<YOUR_TEAM_ID>
MEMBER_ID=<YOUR_MEMBER_ID>
DISH_EMAIL=you@example.com
DISH_PASSWORD=<YOUR_PASSWORD>With that setup:
The MCP will create DISH_COOKIE automatically if it is missing.
If the API returns 401 or 403, the MCP will log in again headlessly, update DISH_COOKIE in .env, and retry the request once.
TEAM_ID and MEMBER_ID stay stable in .env; only the session cookie rotates.
This only works if your DiSH account supports direct email/password login. If your login goes through Google, Microsoft, or another SSO flow, there is no silent refresh path in this repo and you will need the browser-based flow below.
The easiest way to get your credentials is to use the included script:
# Install Playwright browser (first time only)
uv run playwright install chromium
# Run the credential retrieval script
uv run src/get_credentials.pyThis will:
Open a browser window for you to log in to DiSH
Automatically detect when you've reached the dashboard
Capture your cookie, team ID, and member ID
Save them to your .env file
If the automatic method doesn't work, you can retrieve credentials manually:
connect.sid cookieLog in to DiSH in your browser.
Open Developer Tools (F12 / Cmd+Option+I).
In Application/Storage > Cookies, select the DiSH domain.
Copy the connect.sid value (looks like s%3A...).
In your .env file, set DISH_COOKIE to connect.sid=<value>.
Log in to DiSH in your browser.
Open Developer Tools (F12 / Cmd+Option+I).
In the Network tab, find a request to occurrences (or booking-policy).
In the request URL or payload, look for the team and member values.
In your .env file, set TEAM_ID and MEMBER_ID to the respective values.
Keep this secret. Do not commit cookies, team IDs, member IDs, or .env files to source control. Regenerate the cookie if it stops working or was ever exposed.
For use with Cursor or Claude Desktop:
uv run fastmcp run src/mcp_server.pyFor remote access or web-based clients:
# Default: http://127.0.0.1:8000
uv run python src/mcp_server.py --transport http
# Custom port and host
uv run python src/mcp_server.py --transport http --port 3000 --host 0.0.0.0Or using environment variables:
MCP_TRANSPORT=http MCP_PORT=8000 uv run python src/mcp_server.pyThe HTTP server exposes an SSE endpoint at http://<host>:<port>/sse.
"Dish MCP": {
"command": "<PATH_TO_VENV>/bin/fastmcp",
"args": ["run", "<PATH_TO_REPO>/src/mcp_server.py"],
"cwd": "<PATH_TO_REPO>",
"transport": "stdio"
}Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"Dish MCP": {
"command": "<PATH_TO_VENV>/bin/fastmcp",
"args": ["run", "<PATH_TO_REPO>/src/mcp_server.py"],
"cwd": "<PATH_TO_REPO>"
}
}
}The server loads auth values from .env automatically. Keep .env as the source of truth for DISH_COOKIE, TEAM_ID, MEMBER_ID, DISH_EMAIL, and DISH_PASSWORD instead of hardcoding those values in your Claude config.
Book a room using the Dish Manchester API.
Args: datetime_range: Datetime range for the booking meeting_room_name: Name of the meeting room user_info: User information. If not provided, looks for TEAM_ID and MEMBER_ID env vars. cookie: Authentication cookie. If not provided, looks for DISH_COOKIE env var. summary: Title of the booking: default to "meeting"
| Name | Required | Description | Default |
|---|---|---|---|
| datetime_range | Yes | ||
| meeting_room_name | Yes | ||
| user_info | No | ||
| cookie | No | ||
| summary | No | Fuzzy Labs Meeting |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions authentication via cookie and user info with fallbacks to env vars, but lacks details on permissions, error handling, rate limits, or what happens on success/failure. For a mutation tool with zero annotation coverage, this is insufficient.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded with the core purpose. The Args section is structured but could be more concise by integrating details into the main text. Every sentence adds value, though minor streamlining is possible.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 5 parameters with 0% schema coverage and no annotations, the description partially compensates with parameter semantics. However, as a mutation tool with an output schema (not detailed here), it lacks behavioral context like success criteria or error cases. It's adequate but has clear gaps in usage and transparency.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It adds meaningful semantics for all parameters: 'datetime_range' specifies its purpose, 'meeting_room_name' clarifies it's for naming, 'user_info' and 'cookie' explain fallback mechanisms, and 'summary' notes the default. This goes beyond the bare schema, though some details like format specifics are missing.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with a specific verb ('Book') and resource ('a room'), and it specifies the API ('Dish Manchester API'). This distinguishes it from sibling tools like 'cancel_booking' and 'check_availability_and_list_bookings' by focusing on creation rather than cancellation or checking.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus its siblings. It does not mention alternatives like 'check_availability_and_list_bookings' for verifying room availability before booking, nor does it specify prerequisites or exclusions, leaving the agent without context for tool selection.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Cancel a room booking using the DiSH API.
To cancel a booking, you need to know the booking ID. You can get the booking ID by using the
check_availability tool to get the bookings for a room and date range.
Args: booking_id: The ID of the booking to cancel (e.g., "692192791c60f69c20311db3") cookie: Authentication cookie. If not provided, looks for DISH_COOKIE env var. skip_cancellation_policy: Whether to skip cancellation policy (default: False)
| Name | Required | Description | Default |
|---|---|---|---|
| booking_id | Yes | ||
| cookie | No | ||
| skip_cancellation_policy | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
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 discloses that authentication is needed via a cookie or environment variable and mentions a cancellation policy option, which adds useful context. However, it doesn't describe the tool's behavior in terms of side effects (e.g., whether cancellation is reversible, confirmation messages, or error handling), leaving gaps for a mutation tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded with the main purpose. The Args section is structured clearly, though it could be slightly more concise by integrating the booking ID sourcing into a single sentence without repetition.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (a mutation with authentication and policy considerations), no annotations, and an output schema present, the description does a good job covering usage and parameters. However, it lacks details on behavioral outcomes (e.g., success/error responses), which the output schema might address but isn't described here.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema description coverage, the description compensates well by explaining all three parameters: booking_id (with an example ID and how to obtain it), cookie (authentication details and fallback), and skip_cancellation_policy (default behavior). This adds significant meaning beyond the bare schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the specific action ('cancel') and resource ('room booking') using the DiSH API. It distinguishes this tool from its siblings (book_room and check_availability_and_list_bookings) by focusing on cancellation rather than creation or checking availability.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit guidance on when to use this tool: when you need to cancel a booking and have the booking ID. It also specifies how to obtain the booking ID (using the check_availability tool), offering a clear alternative for that prerequisite step.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Check room availability for the DiSH API and list bookings for a room and date range.
Args: datetime_range: Datetime range for the query resource_ids: Optional list of room resource IDs to query. cookie: Authentication cookie. If not provided, looks for DISH_COOKIE env var. ctx: Context object.
Returns: str: The availability summary.
| Name | Required | Description | Default |
|---|---|---|---|
| datetime_range | Yes | ||
| resource_ids | No | ||
| cookie | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions authentication via cookie/env var and returns an availability summary, but lacks critical details: whether this is a read-only operation, potential rate limits, error conditions, pagination for large result sets, or what happens when resource_ids is null. For a query tool with no annotation coverage, this leaves significant behavioral gaps.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is reasonably concise with clear sections (purpose, args, returns). However, the 'Args' section could be more integrated with the main description rather than appearing as separate documentation. Some sentences like 'If not provided, looks for DISH_COOKIE env var.' are useful but could be more elegantly integrated.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 3 parameters with 0% schema coverage and no annotations, the description provides basic context but has significant gaps. It mentions authentication and return type, but doesn't explain the availability summary format, error handling, or query constraints. The existence of an output schema helps, but the description should provide more operational context for a tool with multiple parameters and no annotations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It provides some parameter context: datetime_range is for the query, resource_ids are optional room IDs, cookie is for authentication with fallback to env var, and ctx is a context object. However, it doesn't explain the DatetimeRange structure (start/end datetime), format expectations, or what happens when resource_ids is null. The description adds value but doesn't fully compensate for the schema coverage gap.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Check room availability for the DiSH API and list bookings for a room and date range.' This specifies both the action (check availability and list bookings) and the resource (room). However, it doesn't explicitly differentiate from sibling tools like book_room or cancel_booking, which are clearly different operations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. While it mentions checking availability and listing bookings, it doesn't specify scenarios where this is appropriate versus using book_room or cancel_booking. The only usage context is implicit through the tool's name and purpose.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0book_roomcancel_bookingcheck_availability_and_list_bookingsScored across 3 tools
Each tool has a clearly distinct purpose with no overlap: book_room creates bookings, cancel_booking removes them, and check_availability_and_list_bookings queries availability and existing bookings. The descriptions explicitly differentiate their functions, and the cancel_booking tool even references check_availability_and_list_bookings as a way to find booking IDs, showing complementary rather than overlapping roles.
The tool names follow a consistent verb_noun pattern (book_room, cancel_booking, check_availability_and_list_bookings), all using snake_case. However, check_availability_and_list_bookings is longer and combines two actions, which slightly deviates from the simpler verb_noun style of the others, but the overall pattern remains clear and readable.
With 3 tools, this server is well-scoped for its purpose of managing room bookings via the DiSH API. Each tool earns its place by covering essential operations: booking, canceling, and checking availability/listings. This count is appropriate for a focused domain without being too thin or bloated.
The tool set provides complete CRUD/lifecycle coverage for room bookings: create (book_room), read (check_availability_and_list_bookings), and delete (cancel_booking). A minor gap is the lack of an update tool for modifying existing bookings, but agents can work around this by canceling and rebooking. The domain is clearly covered for core workflows.
AI-native scheduling and booking: check availability, book meetings, share links.
Discover and book businesses via AI agents.
AI-native scheduling: check availability, book meetings, cancel and reschedule via MCP
Scheduling for AI assistants: bookings, availability, meeting polls and routing forms.
