Skip to main content
Glama

ask

Ask the human a question on WhatsApp and BLOCK until they reply (or timeout).

    `text` is the question. `buttons` (optional, up to 3) render as tappable reply
    buttons; omit them to invite a free-text answer the human types on their phone —
    use this to let the human steer you mid-task. `timeout` is seconds to wait
    (max 90). Returns the human's reply (free text and/or the chosen option). On
    timeout, returns a note — the question was delivered and the answer is still
    retrievable via `check_replies`. `to` (optional): "me" (default), "team" or a list
    of team member names or ids; to a group, the first answer wins and names who.
    `mode` (optional, a group): "first" (default) or "all" (every recipient's answer).
    `escalate` (optional, mode "first"): while nobody answers, ask the next people: up
    to 3 steps {"to": "team" or [member names/ids], "after_s": seconds after the send,
    >= 60, each >= 60 after the previous}, or a policy name (POST /v1/team/policies).
    The first answer stops it; it goes on after `timeout`.
    

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": {
      "text": "<text>"
    },
    "name": "ask"
  }
}

Input Schema

NameRequiredDescriptionDefault
toNome
modeNo
textYes
buttonsNo
timeoutNo
escalateNo

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changed
    • addedInput schema / properties / escalate
      Added value: +{
      +  "anyOf": [
      +    {
      +      "items": {
      +        "additionalProperties": true,
      +        "type": "object"
      +      },
      +      "type": "array"
      +    },
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Escalate"
      +}
    • addedInput schema / properties / mode
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Mode"
      +}
    • addedInput schema / properties / to
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    }
      +  ],
      +  "default": "me",
      +  "title": "To"
      +}
  2. First observed

TDQS

A4.5/5.0
Behavior4/5

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

With no annotations, the description carries the full burden and does so well: it discloses blocking semantics, a hard timeout ceiling (max 90s), what happens on timeout (a note is returned and the answer remains retrievable via check_replies), group resolution rules (first answer wins, and it names who), and escalation sequencing. It omits prerequisites such as whether a connected WhatsApp account or recipient availability is required, and says nothing about delivery failure or rate limits.

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

Conciseness4/5

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

Front-loaded with the blocking behavior, and every clause carries operational information, including the dense escalation paragraph. Some sentences are compressed to the point of being hard to parse (e.g. '`mode` (optional, a group)'), trading readability for density.

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?

An output schema exists, yet the description still summarizes the return shape (free text and/or chosen option, or a timeout note) which is helpful rather than redundant. For a 6-parameter blocking tool with group and escalation modes, nothing an agent needs in order to call it correctly is missing.

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?

Schema description coverage is 0%, so the description must compensate entirely, and it does: it defines `text`, `buttons` (up to 3, tappable, omit for free text), `timeout` (seconds, max 90), `to` ("me" default, "team", or member names/ids), `mode` ("first" vs "all", group-only), and `escalate` (up to 3 steps with a >=60s floor per step, or a policy name). This adds meaning far beyond the bare schema types.

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?

States a specific verb and resource ('Ask the human a question on WhatsApp') plus the defining behavioral trait ('BLOCK until they reply'), which immediately separates it from check_replies (non-blocking retrieval) and notify (one-way). An agent can pick this tool 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 Guidelines4/5

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

Gives a clear use case ('use this to let the human steer you mid-task') and points to check_replies for the timeout case, which is genuine routing guidance. It never states the inverse condition — e.g. use notify for a one-way heads-up, or avoid when blocking is unacceptable — so the exclusion side is left to inference.

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.