Skip to main content
Glama

codex_settings

Destructive

Read and update Codex task settings—model, effort, approval and collaboration modes, service tier, access, and family budgets—using operation keys for safe retries.

Instructions

get reads settings. next takes model/effort/approval_mode/collaboration_mode/service_tier/access and root-only family budget (applies now; {} clears). current takes model/effort/approval_mode/service_tier and expected_turn_id. Changes need an operation key.

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>"
    },
    "name": "codex_settings"
  }
}

Input Schema

NameRequiredDescriptionDefault
modelNo
scopeNoget
accessNo
budgetNoFamily lifetime limits: server turns/child admissions hard; active seconds, tokens, stored bytes soft (may overshoot). Native continuations/children may exceed admission caps. {} clears.
effortNo
task_idYesTask id
request_idNoKey for identical retries
operation_idNoAlias of request_id
service_tierNo
approval_modeNo
expected_turn_idNo
collaboration_modeNo

Output Schema

NameRequiredDescriptionDefault
okYes
result_versionYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.2.3

TDQS

A3.6/5.0
Behavior4/5

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

Annotations already declare destructive=true, readOnlyHint=false and non-idempotent, and the description adds real context on top: 'applies now; {} clears' explains that budget writes take effect immediately and how to unset them, and 'Changes need an operation key' discloses a mandatory retry/keying requirement not stated anywhere in the schema. It still omits what exactly is destroyed by 'next' or what permissions are needed beyond root-only budget.

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?

Three short, front-loaded clauses plus a closing requirement, with no filler. The telegraphic style ('next takes ...', 'current takes ...') is dense and slightly ambiguous but every sentence carries information.

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

Completeness3/5

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

Return values need not be explained because an output schema exists, and the mode/parameter mapping plus operation-key rule cover the core flow. For a destructive, open-world, 12-parameter mutation tool, though, the description leaves too much implicit: no statement of what state 'next' overwrites, no value guidance for effort/service_tier, and no prerequisite detail beyond the root-only budget note.

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?

With only 33% schema description coverage across 12 parameters, the description carries a real burden and partially meets it by mapping parameters to scopes (model/effort/approval_mode/collaboration_mode/service_tier/access for 'next', a subset plus expected_turn_id for 'current') — information the schema does not encode. Meaning is still missing for effort, service_tier, access, collaboration_mode and expected_turn_id, so the compensation is partial.

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 description names the three modes explicitly (get/next/current) and states what each one reads or writes, so an agent can distinguish the sub-operations within this one tool. It does not differentiate codex_settings from nearby siblings such as codex_capabilities or codex_goal, and 'get reads settings' is the weakest of the three clauses.

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 is implied indirectly: the per-scope parameter lists tell an agent which scope accepts which arguments, and 'root-only family budget' signals a permission gate. However there is no explicit when-to-use/when-not guidance and no mention of alternatives for reading or changing settings.

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