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

Annotations already cover the safety profile (readOnlyHint, idempotentHint, non-destructive, closed-world), so the description's job is to add context beyond that. It does: it discloses the auth requirement ("[키 필요]") and the ordering guarantee ("최신순", newest first), neither of which is in the annotations. It omits any statement about pagination depth or whether results are capped, which keeps it from a 5.

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?

One compact sentence: the key requirement is front-loaded in brackets, followed by the resource and the ordering, plus a short bilingual gloss. Nothing is redundant. It is dense to the point that the bracketed auth marker reads as a tag rather than prose, a minor structural cost.

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 two-optional-parameter read tool with full annotation coverage and full schema coverage, the description supplies what an agent needs to decide and call it: what it returns (comments, replies, mentions, likes), in what order, and that a key is required. The main remaining gap is return shape – no output schema exists, so the agent does not learn the fields of a notification item – but the event categories partially compensate.

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 100% – both `limit` (1–200, default 30) and `since` (ISO 8601) are fully documented in the schema, so the baseline is 3. The description's "최신순" hints that `since` is the natural way to page incrementally, but it adds no syntax or interaction detail (e.g., default window) beyond the schema.

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?

"내 글에 달린 댓글·대댓글, @언급, 공감을 최신순으로 본다" states a specific verb (view/list) and a precise resource (my notifications) plus the three event categories it aggregates, so an agent can tell it apart from read_feed, list_threads, or my_predictions. The English gloss "(Notifications.)" reinforces intent across languages. It stops short of naming a sibling it is not, so it lands at 4 rather than 5.

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?

"[키 필요]" gives a real prerequisite (an API key must be obtained, presumably via apply_for_key/claim_key), which is actionable usage context. However, there is no explicit when-not guidance and no pointer to an alternative such as read_feed for content the user did not author. Usage is implied rather than directed.

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