Skip to main content
Glama

Delete account metadata

delete-account-metadata
DestructiveIdempotent

Delete one of the account's own metadata field definitions.

Use this tool when:

  • The user explicitly asks to delete a metadata field

metadata_uuid comes from list-account-metadata, and the user must have made clear which field they mean - either by naming it, or because it is the field this conversation has been about. Ask which one they mean when that is not clear, offering the candidates by their display name.

Deleting a field also removes every value recorded for it, on every document and template in the account, and none of that can be brought back. So deleting a field the account has recorded values for is refused until the user has confirmed it: the refusal names the field and how many documents and templates hold a value for it - then repeat the call with confirmed_deleting_values_in_use: true once they have confirmed it. Send that only in answer to such a refusal, never as a matter of course.

Put the effect to the user in their own terms and ask before going ahead - "deleting Discount Rate would also remove the value 12 documents hold for it, shall I go ahead?" - rather than quoting the refusal back at them. Name the field you are about to delete either way, and never delete more than the user asked for.

Where the user wants a field to stop being filled in rather than to lose what it already holds, clear its prompt with update-account-metadata instead: that takes the field out of fynk's AI analyses and leaves the values on their documents alone.

Only the account's own fields can be deleted. fynk's system fields are provided to every account and cannot be removed, system_reference fields included - use update-account-metadata to change how one of those is configured for this account.

Deletion is refused while a bulk metadata operation is still running for the field.

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": {
      "metadata_uuid": "<metadata_uuid>"
    },
    "name": "delete-account-metadata"
  }
}

Input Schema

NameRequiredDescriptionDefault
metadata_uuidYesUUID of the metadata field to delete, as returned by list-account-metadata
confirmed_deleting_values_in_useNoOnly in answer to a refusal naming how many documents and templates hold a value for the field: true states that the user has been told what the deletion takes with it and has confirmed it. Leave it out otherwise

Output Schema

NameRequiredDescriptionDefault
removed_valuesNoWhat the deletion took with the field, to report back to the user
deleted_metadataNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.6/5.0
Behavior4/5

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

Annotations already declare destructiveHint=true and idempotentHint=true, yet the description adds substantial context beyond them: deletion wipes every value on every document and template irreversibly, a refusal gate returns counts when values are in use, and a second refusal occurs while a bulk metadata operation runs. It does not describe the success response, but an output schema exists, so that gap is acceptable.

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 action, then structured into trigger, prerequisite, destructive consequences, refusal loop, and alternatives. It is long, and some of the user-facing coaching ('put the effect to the user in their own terms', 'never delete more than the user asked for') is mild padding, but each section still maps to a decision the agent must make.

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?

For a destructive, gated, two-step tool this covers everything an agent needs: target resolution, the confirmation protocol, refusal conditions, and safe alternatives. The output schema covers return values, so nothing material is missing.

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 baseline is 3; the description goes further by stating where `metadata_uuid` comes from and by prescribing precise semantics for `confirmed_deleting_values_in_use` (only in answer to a refusal, never routinely). That procedural constraint on a boolean flag is real value beyond the schema text.

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 ('Delete ... metadata field definition') and scopes it to the account's own fields, immediately separating it from delete-template-dynamic-field and the update counterparts. The first line is enough for an agent to select it correctly.

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

Usage Guidelines5/5

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

Gives an explicit trigger ('user explicitly asks to delete a metadata field'), names the alternative for the adjacent intent (clear `prompt` via update-account-metadata to stop a field being filled in without losing values), and states the exclusion (system fields, including system_reference, cannot be removed). It even specifies how to disambiguate when the target is unclear.

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