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

Annotations only declare mutation and destructiveness (readOnlyHint=false, destructiveHint=true), leaving the description to carry the full behavioral burden — and it does, in detail. It discloses what 'confirm' freezes (adherence snapshot, re-confirm refresh), exactly what 'discard' destroys (review, kept/skipped/changed marks, actual times, reflection-added events), the per-day validity windows, and the undoToken/expiresAt reversibility mechanism.

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?

The description is long, but nearly every sentence carries load-bearing information for a destructive, two-mode tool with no output schema. It is front-loaded with the core purpose and flows logically from confirm behavior to discard behavior to constraints to reversibility; minor structural formatting (e.g., separating the two actions) could improve scannability without adding content.

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 tool with two divergent behaviors, destructive side effects, and no output schema, the description is remarkably complete: it names the key return fields (`updated`, `deletedIds`, `undoToken`, `expiresAt`), downstream effects (get_schedule's `review` block, stats), prerequisites, validity restrictions, and how to undo. An agent has everything needed to call it correctly for either action.

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 only 50% (the `action` enum has a brief description, `date` has only type+pattern), so the description must compensate — and it does thoroughly. It confirms the date format and adds the past-day restriction, and it expands each action value far beyond the schema ('confirm' freezes a snapshot, 'discard' resets marks and deletes reflection-added events), giving the agent real semantic understanding of both parameters.

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 opening sentence states a specific verb-resource pair ('Confirm or discard a whole day's reflection by `date`'), and the two action values are each given distinct, concrete meanings. It is easy to differentiate from siblings like delete_events (per-event deletion) and confirm_schedule (schedule-level confirmation) because the scope is clearly the day-level reflection state.

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?

The description gives an explicit prerequisite ('once you've marked its events with write_events' `reflect` op'), concrete guidance for choosing confirm vs discard ('use it to start a day's reflection over or drop one confirmed by mistake'), and a hard constraint ('Only a past day can be confirmed; discard works on any day'). It also names the downstream consumers (get_schedule, stats) and the reversal path (undo), so an agent knows exactly when and how to invoke it.

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