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

Annotations already establish readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds a genuinely useful behavioral trait beyond that: 'Deterministic counts, no model,' telling the agent results are reproducible aggregates rather than generated text. It stops short of saying anything about volume, pagination, or freshness of the underlying sessions.

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?

It is front-loaded with an imperative, then the content summary, then a next-step pointer — a sensible order with no filler sentences. Some of the metaphorical phrasing ('already paid for', 'walled ground') spends words on flavor rather than precision, but nothing is redundant.

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 parameters, an output schema covering return values, and annotations covering safety and world scope, the description only needs to convey purpose and sequencing, which it does. An agent could call it correctly from this text alone; only minor details about result size or recency are absent.

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?

The tool takes zero parameters, so there is nothing for the description to disambiguate; the baseline for a parameterless tool applies. The description correctly implies no filtering or scoping arguments are needed before calling.

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 names the concrete content returned: first-pass rules plus live patterns such as refusal routes, stop reasons, and platform issues by surface. That is specific enough to separate it from siblings like get_scoreboard or list_dead_ends. The framing language ('rules already paid for', 'the most walled ground') is evocative but slightly fuzzy about what an individual record looks like.

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 opening imperative 'LEARN FROM EARLIER SESSIONS BEFORE YOU START' gives a clear timing cue, and the closing 'NEXT: list_dead_ends for the ground you are about to work on' explicitly chains it to the follow-up tool. There is no explicit when-not guidance or statement that it takes no arguments and needs no prior state.

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