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

Beyond the readOnlyHint/destructiveHint annotations it discloses real behavior: ranking is "last activity first", results are capped at "up to 8", and a non-matching query falls back to "the most recent ones" rather than erroring. It also states what fields come back (lead_id / organization_id / customer_id) in the absence of an output schema. What it does not cover is the empty-result case or whether legal entities are always appended regardless of the query.

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 intent is front-loaded in the first clause and every segment carries information — output shape, ordering, cap, trigger condition, clarification prompt. It is a single very long em-dash-chained sentence, though, so it reads densely and would scan faster split into two or three sentences.

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?

For a one-parameter read tool with no output schema, the description covers what an agent needs: when to call it, what it returns, ordering and result cap, and the disambiguation fallback. The remaining gap is edge-case behavior — empty results, and whether the legal-entity list is unconditional — which is minor but leaves the agent guessing on rare paths.

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 100% and the single parameter is already documented as "A name or part of it", so the baseline applies. The description reinforces the semantics by defining the no-match fallback path, but adds no syntax, format, or matching-rule detail (e.g. substring vs prefix, case sensitivity) beyond the schema.

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 opens with a specific framing — "Whom an invoice can be issued to" — and enumerates the resource classes it returns (leads with contacts, client organizations, customers already invoiced) plus legal entities. It implicitly separates itself from invoice_list (invoices) by tying output to "the ids invoice_create takes", but it never names a sibling such as erp_list_contacts, so an agent comparing the two list tools gets only indirect differentiation.

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?

It gives an explicit trigger condition: "Call it BEFORE invoice_create whenever the person did not clearly name the client", with a concrete example phrase in Ukrainian. It even prescribes the clarifying question to ask ("на кого виставити рахунок?"), which removes essentially all inference about when this tool belongs in the flow.

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