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

Annotations declare destructiveHint=true and openWorldHint=true; the description earns credit by defusing that ('Nothing goes live or spends without browser approval') and by disclosing the credit cost and the fact that UI cards are returned as structured content rather than rendered. It stops short of saying what the call itself persists or that retries must reuse an idempotency key, which matters given 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.

Is the description appropriately sized, front-loaded, and free of redundancy?

Sentences are short, imperative and front-loaded: purpose, then input rule, then the ask/answer loop, then the guardrail, then the approval outcome. Only the trailing MCP/credits note and the repeated confirmation framing are slightly redundant.

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

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

For a large multi-field brief builder with no output schema, it covers the interaction loop, the prohibition on inventing account/budget/mode values, and the approval gate before spend. What it omits is the retry/idempotency contract and a clearer statement of what is saved between calls – both relevant given the destructive and non-idempotent annotations.

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

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

Schema coverage is 75% and nested, so the baseline is already high; the description still adds real meaning by stating the rule for the unlabeled 'input' object (only confirmed choices, never guesses), the verbatim userStatement requirement, and 'map names only to supplied IDs'. That is a semantic constraint the schema alone does not encode.

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

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

States a specific verb and resource ('Prepare or continue a Meta launch brief') and immediately scopes it to force-collecting a userStatement plus only confirmed input choices. An agent can tell this apart from campaign_drafts, campaigns_setup or editor_settings_reference without opening the schema.

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

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

Explicit ordering rules ('Call before asking questions, then ask the returned open questions ... and call again with the answers') plus hard exclusions ('Never choose an account, budget or launch mode for the user'). It even names the downstream handoff (campaigns_launch) and the free-text fallback, leaving nothing to inference.

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