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

Annotations already cover the safety profile (readOnlyHint, destructiveHint=false, closed-world), so the description's job is to add output semantics — and it does, disclosing that rejected requests carry their reason and that approved requests expose the still-unordered remainder counted as planned spend. That is genuinely non-obvious behavior an agent could not get from the schema. It stops short of stating scope/defaults (does an unfiltered call return all states?) or ordering, which prevents a 5.

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 sentences, no filler, and the most decision-relevant information (what states appear and the unordered-remainder nuance) is up front. The opening is a noun phrase rather than an action statement, which slightly delays clarity about the tool being a listing operation.

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 and only two optional parameters, the description carries the burden of describing return shape, and it does so partially — it explains the unordered-remainder concept but not the overall record fields, volume, or ordering. It is adequate for a low-complexity list tool but leaves real gaps for an agent planning a call.

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 50%: `limit` only says "Default 50" and `state` has no schema text, so the description does partial work by expanding the enum into prose (draft, awaiting approval, approved, rejected-with-reason, ordered). However it omits `cancelled` entirely and never explains `limit` or pagination, so an agent reading only the description would have an incomplete picture of the filter surface.

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 resource (purchase requests / "Requests to spend") and enumerates the lifecycle states it reports, so an agent can tell it returns a status list rather than a single record or a creation action. The verb is only implied (the title says "List"), and nothing distinguishes it from lookalike siblings such as purchase_invoice_list or fin_recurring_list, which keeps it out of 5 territory.

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?

There is no explicit when-to-use statement, no indication of when a different finance or invoice listing tool would be the better call, and no mention that the `state` parameter is the way to narrow to one status. The state enumeration hints at filtering but the agent must infer it.

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