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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, yet the description still adds substantial non-obvious behavior: newest-first ordering, bookmarks:read permission requirement, conditional Your Rec Score field tied to personal recommendation permission, matched_count as the total for applied filters, and the warning not to infer watch history or likes from a save. Those are exactly the traps an agent would otherwise fall into.

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 correctly with the intent list, but the body becomes a dense run-on that mixes retrieval mechanics, pagination policy, response-field inventory, and downstream recommendation etiquette in one block. Most sentences earn their place, yet the volume of procedural advice ("show remaining fetched titles before fetching more...") makes it heavier than a tool definition needs to be.

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 single nested-request parameter with an output schema present, this covers everything an agent needs: filtering, paging, permission prerequisites, total-count semantics, and explicit negative constraints. The return-field list is redundant with the output schema but harmless and does not leave a functional gap.

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?

The schema's own field descriptions are limited (kind has none, and top-level coverage reads as 0%), so the description compensates: query "searches the full list by title fragment", kind "filters movies or shows", and offset is expressed in terms of the returned next_offset cursor. limit is left to the schema, which is the only real gap.

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 — "Lists the connected viewer's own saves, newest first" — and enumerates the concrete user intents it serves (my bookmarks, watchlist, saved titles, "did I save this title"). The phrase "own saves" plus "Cannot read another account" cleanly separates it from sibling tools like search_titles and get_my_ratings.

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 clear when-to-use triggers (browse my list, verify a save, pick from my list) and scope exclusions (cannot read another account, cannot add/remove saves), which implicitly routes write or cross-account needs elsewhere. It also nudges toward check-watch-options behavior. It stops short of naming the specific alternative siblings, so it is strong but not fully explicit about the alternative.

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