Skip to main content
Glama

Approve or reject one inbox item

approvals_decide
Destructive

[PUBLISHES to social accounts · publish scope] Decides ONE item from approvals_list (id): decision "approve" or "reject", optional note. ALWAYS show the user the item's summary (e.g. "Monthly ad spend cap: $500.00 → $1,200.00") and get a clear yes before approving. Ad budget requests ("ad_budget") need approvals:decide: approving a cap change applies it at once; approving any other request lets the agent repeat its exact call once within 7 days. If the workspace requires money approvals to be restated, the answer is 409 confirmation_required: repeat with confirmAmountMinor equal to amountMinor (null for "remove the limit"), only after the user confirmed that amount. Held content works as in decide_approval (approve or approvals:decide; a workflow approval may resume a run that publishes). Repeating the same decision is safe (alreadyDecided: true). Every decision is audited with who, when and that it came through MCP. (Wraps POST /api/v2/approvals/{id}.)

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": {
      "id": "<id>",
      "decision": "<decision>"
    },
    "name": "approvals_decide"
  }
}

Input Schema

NameRequiredDescriptionDefault
idYesFormat: string; e.g. "id_123". An approval id.
noteNoFormat: string; e.g. "Example". On reject: the reason. Optional unless the org requires one.
decisionYesFormat: string; e.g. "approve". Review decision.
confirmAmountMinorNoFormat: integer or null; e.g. 0. Ad budget requests only: restate `amountMinor` (cents; null when it is null) when the workspace requires money confirmation.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changed
    • changedInput schema / properties / confirmAmountMinor / description
      Previous value: -"Ad budget requests only: restate `amountMinor` (cents; null when it is null) when the workspace requires money confirmation."New value: +"Format: integer or null; e.g. 0. Ad budget requests only: restate `amountMinor` (cents; null when it is null) when the workspace requires money confirmation."
    • addedInput schema / properties / decision / description
      Added value: +"Format: string; e.g. \"approve\". Review decision."
    • changedInput schema / properties / id / description
      Previous value: -"An approval id."New value: +"Format: string; e.g. \"id_123\". An approval id."
    • changedInput schema / properties / note / description
      Previous value: -"On reject: the reason. Optional unless the org requires one."New value: +"Format: string; e.g. \"Example\". On reject: the reason. Optional unless the org requires one."
  2. Added

TDQS

A4.6/5.0
Behavior5/5

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

Annotations already declare destructive=true, openWorld=true and readOnly=false, yet the description still adds substantial behavior: approving an ad-budget cap change applies it immediately, other approvals let the agent repeat its exact call once within 7 days, held content can resume a publishing workflow, and every decision is audited with actor, time and MCP origin. The 409 `confirmation_required` protocol and its exact remediation field are documented, which the annotations cannot convey. The note that repeating the same decision is safe (alreadyDecided: true) is scoped to identical decisions and reads as a clarification of the retry edge case rather than a flat contradiction of idempotentHint=false.

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-loaded with the side effect and required scope, followed by the safety rule, then the error path. Nearly every clause carries information, though the parenthetical examples and the inline reference to a `decide_approval` tool make some sentences denser than necessary.

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?

With no output schema available, the description still discloses the meaningful response signals (409 `confirmation_required`, `alreadyDecided: true`) and the audit side effect. For a destructive, open-world mutation tool this covers everything an agent needs to invoke it safely.

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; the description exceeds it by explaining the trigger and required value for `confirmAmountMinor` (must equal `amountMinor`, null when the limit is being removed) and by tying `note` to confirmation flow. That is genuine semantic guidance the schema does not supply on its own.

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 specific verb and resource ('Decides ONE item from approvals_list') and names the accepted values for the decision field. It is immediately distinguishable from the sibling approvals_list (which enumerates rather than acts) and the scope note at the front adds operational context, not noise.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete preconditions: always show the user the item's `summary` and get an explicit yes before approving, and only send `confirmAmountMinor` after the user confirms that amount. It also explains the retry path when a 409 `confirmation_required` is returned. It stops short of comparing against a close alternative (e.g. a bulk or batch decide tool), but the when-to-use context is otherwise explicit.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources