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

Annotations flag destructiveHint=true and idempotentHint=true, but the description adds the operational meaning behind them: removal immediately re-enables serving and can increase spend, unmatched keywords go to not_found with no action taken, and when none match "nothing is sent" (the idempotent/no-op case). It also discloses the matching rules (case- and whitespace-insensitive text, exact match_type required), which the agent cannot infer from annotations.

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?

Front-loaded with the action, then the inverse relationship, then scope limits, then Args/Returns — a logical order with no filler sentences. It runs slightly long, and the Returns block restates information already carried by the output schema, so it is efficient but not maximally tight.

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 destructive, open-world mutation tool with an output schema and full annotation coverage, the description supplies everything an agent needs: parameter formats, match semantics, no-op behavior, partial-failure reporting via not_found, and the spend consequence. Nothing material is left to inference.

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 description coverage is 0% and the keywords array is typed only as a loose object with additionalProperties, so the description carries the full burden — and it does: customer_id is "numbers only, no hyphens," campaign_id is "digits only," and each keyword object requires text plus a match_type enum (EXACT, PHRASE, BROAD) defaulting to BROAD. This is meaning the schema does not supply anywhere.

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 ("Takes negative keywords back out of a campaign") and immediately bounds scope: "Only the campaign's negative keywords are looked at: a keyword it bids on, an excluded location or any other setting is never changed." That scope sentence is what separates it from remove_account_negatives and remove_from_negative_keyword_list without the agent opening any 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?

Gives a concrete routing cue by calling itself "the inverse of add_negative_keywords: pass the same keywords (text and match type)," which tells the agent when to reach for it versus its add counterpart. It also warns about the consequence of use ("can raise what a running campaign spends"). It stops short of explicitly naming the other removal siblings (remove_account_negatives, remove_from_negative_keyword_list) as alternatives.

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