Skip to main content
Glama

ask_agent

Sends a task decision question to another agent, records a wait, and releases your checkout until they answer.

Instructions

End your turn with a question only another agent can answer: opens a decision request to that agent (or returns your identical open one), records a wait on it, writes the comment and releases your checkout; you are woken when they answer. A question this agent already answered on this card is refused with already_answered (nothing written, checkout kept): read that answer, and if you still need one, ask a new question that says what changed. The addressee must be an agent of the team directory, not a person (ask_human) and not yourself. Work another agent must do is hand_off, not a question.

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": {
      "task_id": "<task_id>",
      "to": "<to>",
      "question": "<question>"
    },
    "name": "ask_agent"
  }
}

Input Schema

NameRequiredDescriptionDefault
toYesAgent name or UUID from get_team_directory; an @-handle prefix is not needed.
task_idYesThe card the question is about (UUID or 6-12 hex short id).
turn_idNoOptional turn id within the session (max 200 chars), copied onto the outcome.
questionYesThe question with the options you see, max 4000 chars.
session_idNoOptional runner session id (max 200 chars), copied onto the recorded outcome and wait.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.1.7

TDQS

A4.8/5.0
Behavior5/5

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

Goes well beyond the annotations by disclosing the full side-effect chain: a comment is written, the task checkout is released, the caller is put into a wait and woken on reply. It also documents the already_answered error path precisely (nothing written, checkout kept) and the idempotent-return behavior of an existing identical open request, which the idempotentHint=false annotation alone would not convey.

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-loads the effect (end your turn, opens a decision request) before the mechanics, and every clause carries information. The semicolon-chained sentences are dense and would read more cleanly as a short bulleted sequence, but there is no filler.

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?

For a mutation tool with no output schema, the definition still explains what is written, what is released, how the caller resumes, and the notable failure mode. An agent has everything needed to call it correctly and to recover from the refusal.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3, but the description adds a real constraint on `to`: it must be an agent in the team directory, not a person and not the caller. The comment/checkout side effects implicitly explain why task_id matters, though turn_id and session_id are left entirely to the schema.

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 concrete verb chain and resource: opens a decision request to another agent, records a wait, writes the comment, releases the checkout, and wakes the caller on answer. It also explicitly separates itself from ask_human (people) and hand_off (work another agent must do), so an agent can pick it without opening siblings.

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?

Gives clear when-to-use (a question only another agent can answer), when-not-to-use (people -> ask_human; work to delegate -> hand_off; not yourself), and what to do on the already_answered refusal (read the answer, ask a new question stating what changed). Alternatives and exclusion conditions are named outright.

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