Skip to main content
Glama

codex_delegate

Destructive

Delegate coding tasks to parent-managed native subagents by supplying task_id and prompt; use request_id or operation_id for retries and expected_turn_id for active parents.

Instructions

Ask a parent to manage native subagents; discover them with codex_subagents. Active parents require expected_turn_id.

Input Schema

NameRequiredDescriptionDefault
promptYes
task_idYesTask id
contractNoVersion 1 guidance: version,id,goal,constraints,deliverables,verification,options,evidence,ttl_seconds (60..604800). No authority.
request_idNoKey for identical retries
operation_idNoAlias of request_id
expected_turn_idNo

Output Schema

NameRequiredDescriptionDefault
okYes
result_versionYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.2.3

TDQS

B3/5.0
Behavior3/5

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

Annotations already declare destructive=true, openWorld=true, and non-idempotent, so the safety profile is largely covered. The description adds one genuinely useful behavioral fact beyond them – that active parents require expected_turn_id – but says nothing about what a delegation mutates, what a reused request_id returns, or what the contract's ttl/limits imply. The contract's "No authority" note is only inside the schema.

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 terse clauses with no filler, and the sibling pointer and precondition are both front-loaded. The compression borders on cryptic ('manage native subagents'), but nothing is wasted.

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?

This is a destructive, open-world tool with six parameters, a nested contract object, and an anyOf requirement, yet the description is a single compressed line. An output schema exists so return values needn't be described, but the preconditions, failure modes, and contract semantics an agent needs before invoking it are missing.

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 coverage is 67%, and task_id/request_id/operation_id are already documented in the schema. The description does add meaning the schema lacks for expected_turn_id (the active-parent condition), but leaves the nested contract object and its ttl_seconds/options semantics entirely unexplained.

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

Purpose3/5

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

"Ask a parent to manage native subagents" gives a verb and a resource, but 'ask' and 'manage' are both vague – it is unclear whether this sends a task, spawns a subagent, or issues a control command. It does distinguish itself from codex_subagents by assigning discovery to that sibling, which helps.

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 points to codex_subagents for discovery and states a conditional precondition (expected_turn_id for active parents), which is real routing guidance. However, it never says when to delegate versus using codex_send or codex_respond, so the main use-case decision is left implicit.

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