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

Annotations are all false or unhelpful (readOnlyHint=false for what is clearly a listing operation), so the description carries the burden — and it delivers: newest-first ordering, exact per-item fields, and the two-tier layout of the delta summary 'kept apart' are meaningful behavioral disclosures beyond the schema. Gaps remain: the readOnlyHint=false mismatch is unaddressed, and it doesn't say whether the summary block respects the filters. The description claims no mutating behavior, so this is not a contradiction, but the annotation signal is misleading for a list tool.

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 and roughly 44 words, with the core action front-loaded in the first five words. The first sentence is dense but information-rich, covering order, fields, and summary layout. It loses the top score only because the second sentence partly duplicates what the schema already states about filters.

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, the description's specification of the return shape — per-item fields plus the per-currency approved/pending delta summary — is essential and mostly sufficient for a list call. Minor gaps remain: whether the delta summary is filter-aware, the default result set with no filters, and the precise meaning of 'net delta' vs 'pending delta.' These are non-blocking given all 6 params are schema-documented.

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 100%, so every parameter is already documented with its own description. The description's 'Filter by reference, status, client, date' merely restates a subset of those properties without adding format hints, defaults, or interaction semantics (e.g., how 'to'/'from' behave without 'limit'). The baseline 3 applies because the schema does the heavy lifting and the description adds no parameter-level value.

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?

Opens with a specific verb+resource ('List change orders') and adds concrete detail: ordering ('newest first'), the fields returned ('status, reference, client, currency and net delta'), and a distinctive output feature (a per-currency APPROVED/pending delta summary 'above them... kept apart'). This makes it clearly distinguishable from siblings like change_order_get, change_order_status, and change_order_invoice_payload, which target single records or mutations.

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 'Filter by reference, status, client, date' sentence implies this is the list/search entry point, but the description never states when to use it versus alternatives or names exclusions (e.g., 'use change_order_get for a single record'). Usage context is evident but left to inference, so it sits at the 'implied usage' level rather than explicit guidance.

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