mac_notifications

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

Annotations only say readOnlyHint=false, destructiveHint=false, openWorldHint=true, idempotentHint=false; the description goes well beyond that with side effects (the menu-bar clock click only happens when the center is closed), the meaning of the two return flags center_opened vs center_was_open, a named failure mode ('an absent id returns notification_not_found'), the re-validation guard against a reordered stack, and an explicit privacy/untrusted-input warning. No annotation is contradicted: mutating modes are consistent with readOnlyHint=false, and the clock-click disclosure is exactly the kind of extra context annotations cannot carry.

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?

Purpose is front-loaded correctly, but the rest is a single dense run-on paragraph of chained parentheticals mixing list-mode, dismiss/act-mode, safety, and privacy guidance. Every sentence carries information, yet the lack of separation (bullets or mode-labeled clauses) makes it harder to parse quickly than its length alone would justify.

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 10-parameter, zero-annotation-detail tool with no output schema, the description supplies the return shape inline ({count, notifications:[...], center_opened, center_was_open}), error behavior, and safety guards, which is close to sufficient. The remaining gaps are the undocumented 'limit' and the unresolved name-vs-action distinction.

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?

With 10 parameters and 0% schema description coverage, the description must carry the load, and it does document id, index, app, text, name, expected_app, expected_title and open_center. However, 'limit' is never mentioned, and 'action' is left unaddressed while the prose attaches the action semantics to the separate 'name' parameter ('act (name: an action such as Show, Reply, or press)'), which leaves an agent guessing which of the two fields to populate.

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?

The description opens with a specific multi-verb statement of scope: 'List, dismiss or act on notifications visible to accessibility (banners and alerts on screen, and Notification Center while it is open)'. The scope qualifier ('visible to accessibility') tells the agent exactly what population of notifications is covered, and no sibling tool touches notifications, so there is no ambiguity to resolve.

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?

Mode selection is well covered: 'open_center:true checks whether Notification Center is already open and clicks the menu-bar clock only when closed', and dismiss/act require 'a non-empty id from list (or a 1-based index plus non-empty expected_app / expected_title)'. What is missing is an explicit when-not or a statement about which sibling to use for adjacent tasks, so it falls short of a full 5.

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