Skip to main content
Glama

delegate_task

Queue a bounded worker task for explore, review, or execute work. Read-only modes inspect code; execute runs in an isolated git worktree and may require approval.

Instructions

Queue one bounded worker task. Explore/review are read-only; execute uses an isolated git worktree and may require approval.

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": {
      "objective": "<objective>",
      "workspace": "<workspace>"
    },
    "name": "delegate_task"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindNo
modeNo
nameNo
modelNo
budgetNo
gatewayNo
harnessNo
profileNo
root_idNo
objectiveYes
parent_idNo
workspaceYes
context_packNo
dependenciesNo
allowed_pathsNo
timeout_secondsNo
reasoning_effortNo
success_criteriaNo
execution_channelNo
prohibited_actionsNo
required_capabilitiesNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.8.0
    • addedInput schema / properties / execution_channel
      Added value: +{
      +  "enum": [
      +    "lightworker_worker",
      +    "native_subagent"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / harness
      Added value: +{
      +  "enum": [
      +    "codex",
      +    "zcode"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / required_capabilities
      Added value: +{
      +  "items": {
      +    "pattern": "^[a-z][a-z0-9_]{0,63}$",
      +    "type": "string"
      +  },
      +  "type": "array",
      +  "uniqueItems": true
      +}
  2. Changed5 schema fields changedv0.2.0
    • addedInput schema / properties / budget
      Added value: +{
      +  "additionalProperties": false,
      +  "properties": {
      +    "max_attempts": {
      +      "minimum": 1,
      +      "type": "integer"
      +    },
      +    "max_concurrency": {
      +      "minimum": 1,
      +      "type": "integer"
      +    },
      +    "max_escalations": {
      +      "minimum": 0,
      +      "type": "integer"
      +    },
      +    "max_retries": {
      +      "minimum": 0,
      +      "type": "integer"
      +    }
      +  },
      +  "type": "object"
      +}
    • addedInput schema / properties / context_pack
      Added value: +{
      +  "oneOf": [
      +    {
      +      "maxLength": 32768,
      +      "type": "string"
      +    },
      +    {
      +      "additionalProperties": false,
      +      "properties": {
      +        "content": {
      +          "maxLength": 32768,
      +          "type": "string"
      +        },
      +        "name": {
      +          "type": "string"
      +        },
      +        "version": {
      +          "type": "string"
      +        }
      +      },
      +      "required": [
      +        "name",
      +        "version",
      +        "content"
      +      ],
      +      "type": "object"
      +    }
      +  ]
      +}
    • addedInput schema / properties / gateway
      Added value: +{
      +  "type": "string"
      +}
    • addedInput schema / properties / profile
      Added value: +{
      +  "type": "string"
      +}
    • changedInput schema / properties / reasoning_effort / enum
      Previous value: -[
      -  "low",
      -  "medium",
      -  "high",
      -  "xhigh"
      -]New value: +[
      +  "low",
      +  "medium",
      +  "high",
      +  "xhigh",
      +  "max"
      +]
  3. First observedv0.1.1

TDQS

B3.3/5.0
Behavior4/5

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

Annotations only say readOnlyHint=false and destructiveHint=false; the description adds materially useful traits beyond that: the task is queued (asynchronous), 'execute' runs in an isolated git worktree (containment, which justifies the non-destructive hint), and execute 'may require approval' (a gate the agent must anticipate). It stops short of covering failure/retry/escalation behavior.

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?

Two sentences, front-loaded with the core action and then the safety-relevant kind distinctions. No filler, but the brevity is arguably under-specification given the parameter surface rather than genuine economy.

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 21-parameter orchestration tool with no output schema and zero schema-level parameter documentation, the description is far too thin: it never explains what queuing returns (task id?), how dependencies/budget constrain execution, or how approval interacts with 'mode'. Only the kind semantics and isolation model are covered.

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

Parameters2/5

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

Schema description coverage is 0% across 21 parameters, and the description only illuminates 'kind' (explore/review read-only vs execute isolated). Budget, dependencies, allowed_paths, context_pack, success_criteria, prohibited_actions, required_capabilities, harness, and execution_channel are entirely undocumented in both schema and description.

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?

States a specific verb and resource ('Queue one bounded worker task') and distinguishes the three task kinds. The word 'one' implicitly contrasts with the sibling delegate_batch, but the contrast is never made explicit, so an agent still has to infer the boundary.

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?

The description gives real selection guidance at the kind level (explore/review read-only, execute isolated and possibly approval-gated), which helps pick a kind. It never says when to call delegate_task rather than delegate_batch, orchestrate, or native_subagent_started, leaving the tool-level choice to inference.

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