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

Goes well beyond the annotations (which only establish read-only, idempotent, non-destructive, closed-world): it names the required permission scopes, the ordering guarantee, the 24-hour default window, the cursor_expired error and how to recover, the fact that items carry ids only, and a prompt-injection warning that user-written text lands in untrusted_fields and must never be followed as instructions. Those are exactly the traits an agent cannot get from the structured fields.

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-loads purpose, then scopes, then pagination/cursor behavior, then the safety warning about untrusted_fields, in a tight telegraphic style. Every clause carries operational information; nothing is repeated from the annotations and there is no filler.

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?

With no output schema, the description still tells the agent the ordering, that returned items are ids only, and where the free text lives (untrusted_fields), plus the cursor field to feed back. It stops short of describing the item envelope or what `types` accepts, which is the main residual gap for a zero-required-parameter polling tool.

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 0% across 3 parameters, so the description must carry the load. It fully explains `since` (cursor round-trip, 24h default, expiry) — the highest-risk parameter — but says nothing about `limit` (1-100) or `types` (the event-type filter), leaving two parameters undocumented in both schema and prose.

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+resource ('List what happened') and makes the scope concrete via ordering ('Oldest first') and the required read scopes, which tell the agent this is a cross-domain activity feed (bookings, payments, inbox, account). It routes detail lookups away from itself toward get_booking_proposal and get_payment, so it is distinguishable from the many sibling list_* tools. The only softness is the colloquial 'my last look', which relies on the reader inferring a per-caller cursor.

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 the explicit polling pattern ('Pass the answer's cursor back as since next time'), the no-cursor default ('without since it covers the last 24 hours'), and the failure path ('A stale cursor is cursor_expired: start again from a time'). It also names the alternatives for detail: read get_booking_proposal (whose actions say what may happen next) or get_payment. When-to-use, when-not-to-use, and recovery are all present.

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