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

Annotations already declare readOnlyHint, idempotentHint, openWorldHint and destructiveHint=false, so safety is covered; the description adds the streaming pattern (cursor loop) and instructs relaying each event's `say` to the user, which is real behavioral value. However it recommends `wait=30` while the schema caps `wait` at 25 (and the param text says 0-30 s), so following the description literally can produce an invalid call, and it says nothing about termination or end-of-run signalling.

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?

Two short sentences, imperative and front-loaded: the action first, then the alternative-usage rule. No filler or restated title text.

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 carries the burden of explaining the event stream, yet it only mentions `say` — cursor semantics, how `types`/`include` change the payload, and how the tail terminates are left unexplained. Adequate to start a call, insufficient to consume the stream correctly.

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%, so all five parameters are documented structurally; the description only sketches a concrete call shape (`after=<cursor>&wait=30`) and mentions `say` as the payload field to surface. It adds little semantic depth and the concrete value it supplies conflicts with the declared maximum of 25.

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?

"Follow a deep research run step by step" names a specific verb (follow/tail) and resource (a deep research run), and the title adds the transport mechanism (long-poll or SSE). It is clearly distinguishable from siblings like get_deep_analysis, though it never states what an event looks like beyond the `say` field.

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?

"Use instead of polling get_deep_analysis while the run is going" gives an explicit alternative and the condition that selects it. It stops short of saying what to do once the run finishes or the cursor is exhausted, which is the natural complement to that rule.

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