Widely

Create an Errands call

create_call_task

Places or schedules an Errands call from a brief. The price comes from the rate sheet. A future schedule_at waits until then. There is no global calling-hours window. Fill every field you can; call describe_errand_template for the fields and examples. Emergency and premium-rate numbers are refused. A marketing brief is refused. The seat needs the current Errands terms. Business plan. A bad brief lists every missing and invalid field at once. The assistant introduces itself with the name set in the user's Widely AI assistant settings (Assistant's name).

Input Schema

Name	Required	Description	Default
`goal`	Yes	The outcome you want, in one or two sentences.
`intent`	No	Optional. Why you are calling, in one sentence. If you leave it out, the opening line is used, then who the call is for.
`limits`	No
`callback`	No
`language`	No	BCP-47 such as en-GB or he-IL, or auto-from-seat.
`forbidden`	No	Things the assistant must not say or agree to.
`ivr_hints`	No	Phone menu hints, e.g. 'choose imports'.
`voicemail`	No	On voicemail: leave_message (default), hang_up_and_retry (uses limits.retry), or hang_up.	leave_message
`escalation`	No
`if_unknown`	No	What to do when asked something the brief does not cover.	say_will_check_back
`time_limit`	No	Optional. Minutes the assistant has for this call. If you leave it out, limits.max_call_minutes is used when that is set, otherwise 5.
`callee_name`	Yes	Who you are calling, e.g. 'FedEx UK Import Clearance' or 'Dr Cohen's clinic'.
`destination`	Yes	Number to call, E.164 (+44…, +972…, +1…) or local. Emergency and premium-rate numbers are refused.
`schedule_at`	No	When to call, ISO date-time, at most 14 days ahead. Omit it to call now.
`time_window`	No	Optional calling window, at least 10 minutes long and within 14 days. The call is placed inside it.
`connect_user`	No	When true, the assistant asks the callee to hold at the end, rings the user's phones into the call, and leaves.
`on_behalf_of`	Yes	Who the call is for. person is required.
`opening_line`	No	Optional. What to say first, after the assistant introduces itself and plays the mandatory notice.
`facts_to_give`	No	Facts the assistant may give. Card numbers, CVVs, PINs and passwords are refused.
`assistant_name`	No	Optional. The name the assistant uses on this call instead of the AI assistant settings name.
`save_as_default`	No	With assistant_name, also save it as the default name in Widely AI assistant settings.
`client_reference`	No	Your own reference.
`facts_to_collect`	No	Questions to ask and the information to bring back.
`purpose_category`	Yes	Kind of call. Marketing, sales, surveys, debt collection, political and fundraising calls are refused.
`success_criteria`	Yes	How to tell the call worked, e.g. 'they confirm the delivery date'.
`voicemail_message`	No	Optional. If absent, the assistant leaves a short message from the goal and on_behalf_of, with no sensitive facts.
`callee_is_business`	No	Optional. True when the callee is a business.
`caller_identity_id`	No	Optional. You may set it. If you omit it, Widely picks the caller ID. If you set it, it must be an identity_id from list_identities, one of the user's own. Never invent another number.
`allowed_commitments`	No	The only things the assistant may agree to, e.g. 'accept any slot Tue–Thu after 14:00'. For anything else it says it will check and come back, and lists it in the report. Payments, contracts and cancellations are never allowed.

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 26, 2026

addedInput schema / properties / intent

Added value: +{
+  "description": "Optional. Why you are calling, in one sentence. If you leave it out, the opening line is used, then who the call is for.",
+  "maxLength": 300,
+  "type": "string"
+}

addedInput schema / properties / time_limit

Added value: +{
+  "description": "Optional. Minutes the assistant has for this call. If you leave it out, limits.max_call_minutes is used when that is set, otherwise 5.",
+  "maximum": 120,
+  "minimum": 1,
+  "type": "integer"
+}

Changed1 schema field changedOct 25, 2026

changedInput schema / properties / caller_identity_id / description

Previous value: -"Optional identity id. Omit it and Widely picks the caller ID the same way the app does."New value: +"Optional. You may set it. If you omit it, Widely picks the caller ID. If you set it, it must be an identity_id from list_identities, one of the user's own. Never invent another number."

AddedOct 22, 2026

TDQS

A3.8/5.0

Behavior5/5

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

Annotations already declare readOnly=false, openWorld=true, idempotent=false, and destructive=false, and the description adds substantial operational context: price from the rate sheet, schedule_at waiting behavior, no global calling-hours window, refusal rules, terms/plan prerequisites, and aggregated validation errors. This gives the agent meaningful behavioral expectations beyond the structured annotations without 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.

Conciseness3/5

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

The purpose is front-loaded and the overall length is reasonable for a 29-parameter tool. However, the prose is telegraphic and includes ambiguous fragments such as 'Business plan.' that weaken structure and clarity.

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 29 parameters, no output schema, and high schema coverage, the description covers prerequisites, scheduling behavior, refusals, validation behavior, and template guidance well. It still omits what a successful creation returns, such as a task identifier, and does not explicitly route to closely related tools.

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?

Schema description coverage is 90%, so the schema already documents most of the 29 parameters in detail. The description adds useful scheduling context around schedule_at and time_window and advises filling all possible fields, but it does not meaningfully explain most parameter semantics beyond what the schema already provides.

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

Purpose4/5

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

The description opens with 'Places or schedules an Errands call from a brief,' giving a specific verb, resource, and Errands scope. It is clear what the tool creates, but it does not explicitly distinguish this action from siblings such as quote_call_task, request_call, or control_call.

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?

It provides prerequisites and refusal cases ('current Errands terms', 'Business plan', emergency/premium and marketing calls refused) and directs the agent to describe_errand_template for field guidance. However, it never states when to choose this tool over related siblings like quote_call_task or request_call, so usage is mostly implied.

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

All Widely tools Inspect

Try in Browser

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

Glama MCP Gateway

Add one secure layer between your agents and this server.

Create an Errands call

Input Schema

Schema Changelog

TDQS

Glama MCP Gateway

Resources