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

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds real value beyond that: an explicit prompt-injection warning that untrusted_text fields are third-party data, not instructions, and that "⚠ flagged text:" names are suspicious. It also enumerates returned fields and aging buckets and notes the concise-vs-detailed output tradeoff, though it says nothing about pagination behavior.

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 key owed_to_you/you_owe split is front-loaded, which is good, but the first sentence is an overlong run-on enumerating status, totals and aging buckets that largely restate returned data. The safety note is well-placed at the end, but the middle budget spent on output fields costs clarity for an 11-parameter search tool.

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 11 optional params, no output schema, and nested amount object, the description does meaningful work: it defines the side semantics, sketches the return payload, and adds a prompt-injection caveat that no structured field carries. It stops short of explaining pagination (cursor/limit is only valid for single-side queries) and the response_format tradeoff beyond what the schema says.

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 description coverage is only 45%, so the description is expected to compensate and partially does: it explains the side parameter's owed_to_you/you_owe semantics, which the schema enum does not define. But cursor, limit, response_format, due_before, overdue_days_min and the amount-tolerance filter are left to the schema, and the long field list describes outputs rather than the undocumented inputs.

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 (search) and resource (invoices/bills), and draws the owed_to_you vs you_owe distinction plus the field/aging-bucket coverage. It signals it is a lookup that precedes an add/pay action, which leans toward add_invoice. However it never names or distinguishes itself from the sibling query_invoices, which appears to be the closest alternative, so the differentiation is incomplete.

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?

"Finds the existing invoice or bill before one is added or paid" implies a use case (dedupe/verify before add_invoice or record_payment), which is genuine implied guidance. But there is no explicit when-not, and no routing between this tool and query_invoices or search_transactions. An agent must guess which of the two invoice-search tools to use.

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