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

Goes well past the readOnlyHint=true annotation: it discloses the permission gate on company scope, the null transaction_id for company callers and why, that amounts are unsigned magnitudes, that `settlement` is computed at read time and never stored with an explicit 'never re-derive it yourself' instruction, and that success:false is a failure rather than an empty result. That last point is exactly the kind of trap an agent would otherwise fall into.

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-loaded with purpose, then scope, then return shape, then failure semantics, then workspace selection — a sensible order. The long sentence enumerating returned fields is somewhat redundant given an output schema exists, but it carries real semantics (unsigned amount, null transaction_id on company scope), so nearly every clause earns its place.

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?

Covers everything an agent needs to call this correctly despite four optional parameters and two scope modes: default behaviour, permission requirements, the workspace ambiguity rule, the meaning of an empty list, and the failure signal. Because the output schema already defines the payload, the description's field enumeration is a bonus rather than a gap-filler.

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?

At 75% schema coverage the baseline is around 3, and the description earns above that by giving `scope: "mine"` an explicit default, tying `company` to an owner/admin requirement, and restating the workspace_id rule with its rationale ('this read will not guess which one you mean'). `conversation_id` and `workspace_scope` are left entirely to the schema, which is acceptable since those schema descriptions are self-contained.

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 and resource ('List the expense notes of the person who asks') and immediately disambiguates the concept with a plain-language gloss ('payments they made with their own money ... which the company owes them back'). It also splits the resource by scope ('mine' vs 'company'), so an agent knows exactly which slice it is fetching without opening the schema.

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?

Explicit conditions for each mode: default scope 'mine', 'company' scope being read-only for owners/admins, and the workspace_id rule ('when the token authorizes one workspace, call this directly — when it authorizes several ... pass workspace_id'). It does not name the complementary siblings (well_record_expense_note, well_list_expense_groups), so an agent must infer when to record vs list, but the in-tool routing guidance is otherwise strong.

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