Skip to main content
Glama

Read, cancel or reschedule one booking with its cancel token

ic_scheduling_manage_booking

The way OUT of a booking, and it is first-class on purpose: an agent that has to ask its human to click a link in an email simply ghosts, and no-shows are the failure mode that actually burns members' time. NO TOKEN REQUIRED — the booking's own cancel_token IS the credential, exactly as a form's claim token is for an account-less submitter. Putting the exit behind an IC account nobody was issued is how a person ends up emailing a human to be removed. Authorization has not been skipped, it has MOVED into the resource: the token is checked in constant time and a WRONG token gives the SAME answer as a missing booking (not_found), so this cannot be used to discover which booking ids exist. action:read returns the booking's current state. action:cancel is idempotent — cancelling an already-cancelled booking is a SUCCESS, not an error, so a retry after a dropped response does not look like a failure. action:reschedule NEEDS start. A refused move leaves the original booking untouched: the release and the new booking are one transaction, so an error means nothing changed and you may try another slot. Confirm cancel and reschedule with your human first; both are visible to the member immediately. booking.id in a reschedule response is NEW and is the one to keep; cancel_token is unchanged. Reading the OLD id with the same token returns the LIVE booking (booking.id is the new id) plus superseded{id,start,end,rescheduled_by} naming the booking you asked for, so a saved manage link keeps working after the owner moves the meeting. action:read is also how you wait for a Google Meet link: calendar.meet_link.status pending -> ready (location_url is the link) or failed/none (no link is coming); poll no faster than every 2 seconds, for at most 90 seconds. Args: { booking_id, cancel_token, action, start?, reason?, idempotency_key? }. Returns: { ok, booking{id,status,start,end,cancel_token,cancelled_by,rescheduled_by,rescheduled_to}, superseded?{id,start,end,rescheduled_by}, member, meeting_type, location_url, ics, calendar{status,meet_link{status,reason,retrying}} } or { ok:false, error_kind } from not_found | slot_taken | outside_window | too_soon | rate_limited | validation | transient. No auth required.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "booking_id": "<booking_id>",
      "cancel_token": "<cancel_token>",
      "action": "<action>"
    },
    "name": "ic_scheduling_manage_booking"
  }
}

Input Schema

NameRequiredDescriptionDefault
startNoRequired for action:reschedule. An ISO-8601 slot start from a FRESH ic_scheduling_get_availability call — the old response is stale the moment anyone else books.
actionYesread is safe and repeatable. cancel and reschedule change a real person's calendar — confirm with your human first.
reasonNoOptional, for action:cancel. The member sees it; it is the difference between a cancellation and a ghosting.
booking_idYesThe booking's id, from the confirm response's booking.id.
cancel_tokenYesFrom booking.cancel_token on the confirm response, or the manage link inside the .ics your human saved. It is the only credential for this booking.
idempotency_keyNo8-128 chars of [A-Za-z0-9._:-]. Required by the service on a reschedule; minted here if you omit it. Bring your own and reuse it on retry.

Output Schema

NameRequiredDescriptionDefault
okNotrue when the call did what it was asked; false with `error` and `error_kind` when it did not
textNoThe tool's reply when it returns plain text instead of JSON
errorNoWhy the call failed, in a sentence a person can act on. Present only when ok is false.
resultNoThe tool's value when it returns a list or a single value instead of an object
error_kindNoMachine-readable failure class: no_token, missing_scope, forbidden, not_found, invalid_input, rate_limited or transient

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "$schema": "http://json-schema.org/draft-07/schema#",
      +  "additionalProperties": {},
      +  "description": "The tool's JSON result. Every field the tool's description names is present at the top level; ok, error and error_kind follow the shared envelope.",
      +  "properties": {
      +    "error": {
      +      "description": "Why the call failed, in a sentence a person can act on. Present only when ok is false."
      +    },
      +    "error_kind": {
      +      "description": "Machine-readable failure class: no_token, missing_scope, forbidden, not_found, invalid_input, rate_limited or transient"
      +    },
      +    "ok": {
      +      "description": "true when the call did what it was asked; false with `error` and `error_kind` when it did not"
      +    },
      +    "result": {
      +      "description": "The tool's value when it returns a list or a single value instead of an object"
      +    },
      +    "text": {
      +      "description": "The tool's reply when it returns plain text instead of JSON"
      +    }
      +  },
      +  "type": "object"
      +}
  2. Added

TDQS

A4.4/5.0
Behavior5/5

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

The description discloses behavior well beyond the annotations: the cancel token is a credential checked in constant time, wrong tokens and missing bookings both return not_found (preventing booking-id discovery), action:cancel is idempotent so retries after dropped responses succeed, action:reschedule is transactional so a refused move leaves the original untouched, reschedule returns a NEW booking.id and the old id read returns superseded details, and polling limits for Meet link readiness. These details materially change invocation and retry behavior. The per-action note that cancel is idempotent does not contradict the tool-level idempotentHint:false, because reschedule is not idempotent; it clarifies the mixed-action semantics rather than contradicting them.

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

Conciseness2/5

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

The description is roughly 450 words and opens with rhetorical justification ('The way OUT of a booking... simply ghosts... no-shows are the failure mode that actually burns members' time') before stating the actions. It also includes analogies to claim tokens and commentary about emailing a human that, while clarifying design intent, do not directly help an agent invoke the tool. The factual content is valuable and organized later (actions, confirmation, new id, polling, Args, Returns), but the brevity and front-loading are poor.

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

Completeness5/5

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

The tool has an output schema and annotations, and the description still adds the crucial runtime context: error_kind values, polling limits, superseded behavior, transactional guarantees, and the idempotency nuance. Nothing an agent needs in order to call this tool correctly appears missing. The completeness is strong for a multi-action mutation tool with a credential-based authorization model.

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?

The schema has 100% description coverage and already documents each parameter's meaning, including 'Required for action:reschedule' for start, 'Optional, for action:cancel' for reason, and the idempotency_key sourcing/reuse instructions. The description's Args line merely lists names and adds no parameter meaning beyond the schema. The schema, not the description, carries the parameter semantics; baseline 3 applies.

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 names three specific actions (read/cancel/reschedule) on a specific resource (one booking) and explicitly ties them to the cancel token, which is the core identity of the tool. It also distinguishes this tool from ic_scheduling_get_availability by requiring a fresh availability call for reschedule. An agent can tell exactly what this tool does without opening the schema.

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

Usage Guidelines5/5

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

It gives explicit when-to-use guidance for each action: action:read returns current state and is the way to poll for a Meet link, action:cancel is for exiting, action:reschedule is for moving with a fresh start slot. It also states the key precondition (confirm cancel/reschedule with your human first), the polling cadence (no faster than every 2 seconds, at most 90 seconds), and names ic_scheduling_get_availability as the source of a valid start. The only minor gap is no explicit 'when not to use', but the action-level routing is unusually complete.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.