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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds genuinely new behavioral context beyond that: no provider/browser call is made, no scan allowance is consumed, and missing or failed evidence is reported as unavailable rather than zero — a meaningful semantic caveat. It also flags quoted source text as untrusted data. Only return format and pagination behavior are left unstated.

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 body is dense and mostly earns its place, but the lead position is spent on the title restatement 'Saved structured-data comparison', which conveys nothing new. The remaining sentences are compact and information-bearing. Structure is adequate but the front-loading is wasted.

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?

Annotations cover the safety profile, but there is no output schema and the description never explains the shape of the returned comparison or how pagination works, so an agent cannot predict results. The missing-evidence and untrusted-text caveats are valuable, yet the total picture is incomplete for a tool with three undocumented parameters and no return-value description.

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?

There are three parameters (chars, limit, offset) with 0% schema description coverage, so the description carries the full burden of explaining them — and it says nothing about any of them. No guidance on the character budget, result cap, or pagination offset appears anywhere in the text. Conventional names are not enough to compensate for a complete coverage gap on a paginated tool.

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 'Saved structured-data comparison' is essentially a restatement of the title, so the first impression is tautological. The second sentence does add a concrete scope ('reads saved measurements or existing caches only'), which tells the agent this is a read of previously stored data rather than a fresh computation. But it never states what the comparison produces or how it differs from the many other 'get_*' evidence siblings, leaving the purpose only partly resolved.

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 clause 'no provider/browser call or scan allowance is consumed' implies the right situation to call it: when the agent wants saved evidence without spending a rescan budget. That is useful implied guidance, but no explicit when-to-use/when-not framing is given, and no sibling (e.g. get_cited_queries, get_source_profile, request_rescan) is named as the alternative to pick instead. The agent must infer the routing itself.

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