Skip to main content
Glama

run_agent

Run a single subagent task and wait for completion, validating the provider, model, effort, and permissions you specify. Supports read-only or workspace-write limits and conversation continuation.

Instructions

子を1件実行し完了を待つ。permission省略はfull-access(信頼する子向け・OSユーザー権限の範囲)。制限はread-only/workspace-write(実行先が対応する場合)。このOSにbrokeredは無い。指定モデル・Effort・権限を検証する

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": {
      "provider": "<provider>",
      "model": "<model>",
      "task": "<task>"
    },
    "name": "run_agent"
  }
}

Input Schema

NameRequiredDescriptionDefault
cwdNo
roleNo
taskYes
filesNo
modelYes必須。許可済みの取得済みモデルID
effortNoEffort対応モデルでは必須。対応・許可済みの値のみ
contextNo
providerYesプロバイダーID
permissionNo省略時full-access(OSユーザー権限の範囲で標準編集・シェル)。read-only/workspace-writeは既存のCLI標準制限。brokeredではread-only/workspace-writeの明示が必須
timeout_msNo
json_schemaNo
conversation_idNo同じサーバーの完了した会話を継続するID
include_detailsNo同期結果に使用量・実行メタデータを含める。既定false。非同期結果はget_agentで指定
notify_progressNo同期実行の進捗通知を送る。既定false。バックグラウンド実行では通知しない
output_instructionsNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.0.0

TDQS

B3.3/5.0
Behavior3/5

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

With no annotations, the description carries the full burden, and it does disclose two non-obvious behaviors: the call blocks until completion, and omitting permission grants full-access (standard editing and shell within OS user permissions). It says nothing about timeout_ms behavior, what happens on child failure, whether a new agent or existing conversation is used, or what is returned — gaps that matter for a 15-parameter agent-spawning tool.

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?

Four dense sentences, front-loaded with the action and the blocking semantics, with no filler. The brokered remark is terse to the point of being cryptic out of context, but every sentence still carries information.

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

Completeness2/5

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

For a 15-parameter, annotation-free, schema-partial tool that spawns a subagent with shell access, the description omits the return shape (no output schema exists), timeout and failure behavior, and async alternatives. It covers the highest-risk item — the permission default — but leaves too much of the calling contract to guesswork.

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 47%, and the description mostly restates what the schema already says about permission (including the same full-access default) while adding the behavioral note that model, Effort, and permission are validated against allowed values. Roughly half the parameters (task, role, files, cwd, timeout_ms, context, conversation_id, json_schema) get no added meaning, so it does not fully compensate for the coverage gap.

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 first sentence gives a specific verb+object: run a single child agent and block until it completes ("子を1件実行し完了を待つ"). The singularity plus the blocking wait implicitly separates it from parallel/async siblings, but no sibling is named, so an agent must infer the distinction.

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?

Usage context is implied by "完了を待つ" (synchronous execution) and by the note that permission defaults to full-access for trusted children, but there is no explicit when-to-use versus run_agents_parallel, start_agent, or send_agent_message guidance. The one firm directive — that this OS has no brokered mode — is a useful constraint but not selection guidance.

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