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

Annotations already disclose a non-read-only, non-destructive, non-idempotent mutation, and the description adds meaningful traits on top: duplicates are silently dropped, large inputs are batched, and the effect propagates to every attached campaign at once. The propagation blast radius is exactly the kind of context annotations cannot convey, though permission requirements and failure behavior are unstated.

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 behavior, then the sibling/alternative, then structured Args and Returns. The inline SQL example is longer than strictly necessary but earns its place by making shared_set_id discovery concrete.

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 3-parameter mutation with a nested array and zero schema documentation, the description covers what to pass, where the ID comes from, how it behaves at scale, and what is returned. It falls short only on prerequisites (permissions, whether the caller needs list ownership) and on disambiguation from adjacent negative-keyword tools.

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 0%, so the description carries the full burden and does so well: customer_id format (digits only, no hyphens), what shared_set_id refers to, and the nested keyword object with its text field plus the EXACT/PHRASE/BROAD enum and a BROAD default that the schema does not encode. Minor gap: it does not state that text is required or what happens with unexpected keys the loose schema would allow.

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 ('Adds negative keywords to an existing shared negative keyword list') and makes the scope concrete: the words apply to every campaign the list is attached to. It does not, however, distinguish itself from the close siblings add_negative_keywords and add_account_negatives, which an agent could easily confuse with this tool.

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?

Clear operational context: the list must already exist, its ID is discoverable via the search tool (with a concrete query), and removal is routed to remove_from_negative_keyword_list. What is missing is explicit when-not guidance against the campaign-level and account-level negative keyword tools, which is the main selection risk here.

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