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

Annotations only declare the mutation profile (readOnlyHint=false, idempotentHint=true, destructiveHint=false); the description adds substantial context beyond that: a new revision is written, an effect report carries n_changed/rejected/cloned/carried, and partial failure is possible (rejections are per element). It further discloses the exact refusal conditions (target class has other attributes, unknown PredefinedType, feature elements via insert_in_host) and the attribute carry-over rule, which is precisely the behavioural detail an agent needs before mutating a model.

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?

Purpose and preservation semantics are correctly front-loaded, and nearly every clause carries real information. But the prose is telegraphic and under-paragraphed — 'A call that succeeds is not a job that's done' and the trailing type-object sentences ('class being theirs', 'the type takes the type class the schema pairs with it') require re-reading and slow an agent trying to parse the call sequence.

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 an 11-parameter mutation tool with no output schema, the description covers the preconditions, the plan gate, the failure model and the attribute migration rules, which is most of what an agent needs. The gaps are the meaning of the individual effect-report counters (n_changed, rejected, cloned, carried are named but not defined) and the absence of an explicit statement of which parameter combinations are valid versus merely optional.

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 the baseline is 3; the description earns more by explaining cross-parameter relationships the schema documents only in isolation — set paired with expect_count, plan_id paired with expect_beyond, and the dry_run-to-plan_id-into-expect_beyond chain. It also constrains target selection to guid/guids/set and rules out filtering. It stops short of explaining request_id/note/dry_run interplay, keeping it below a 5.

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 opening sentence names a specific verb and resource with a concrete example ('Changes the IFC class (IfcSlab to IfcRoof)') and immediately states what is preserved (GlobalId, geometry, properties, relations). That preservation clause also implicitly separates it from delete_elements + create_elements, so an agent can distinguish it from siblings 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?

It gives explicit targeting guidance ('Target: guid, guids or a set from find_elements/get_selection, never a filter') and an explicit exclusion ('Deleting and creating is no substitute - properties and relations would be lost'). It also mandates a PLAN_REQUIRED dry-run flow before going beyond the named elements. What is missing is any pointer to the sibling that should be used instead for adjacent jobs (e.g. set_properties for attribute-only edits), so the routing is one-sided.

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