cad

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

No annotations are provided, so the description carries the full burden. It discloses key behavioral traits: session step recording and rollback, unified transaction per mutation, connectivity audit, obj_name semantics for new versus base objects, return values (step number, optional screenshot), and batch behavior with stop_on_error. It does not mention what happens when no session is active or permission requirements, but the disclosed traits are substantial.

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 description is moderately sized but well-structured: a one-sentence purpose, followed by context, then a clear Args list, and a Returns paragraph. Every sentence adds information (session behavior, transaction/audit, obj_name rule, operation_help pointer). It is not as succinct as a two-sentence description, but complexity justifies the length.

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?

Given the tool's complexity (9 parameters, 22 operations, no annotations), the description covers the essential contextual pieces: session integration, transaction/audit behavior, the distinction between new-object and base-object naming, and the return format. It wisely points to operation_help for per-operation parameters. It does not explicitly state prerequisites like calling session_start, but implies it by saying 'When a modeling session is active'.

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 0%, so the description compensates by explaining critical semantics: obj_name is a base object except for sketch/variables/datum_plane/hull where it names a new object; obj_properties holds operation parameters; stop_on_error is for batch; description is a step log note; with_screenshot attaches a screenshot. It also directs to operation_help for full parameter details. Some parameters (obj_type, ops) are not explained in the description, but the most nuanced ones are.

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 'Perform a CAD modeling operation (unified mutation tool)', which clearly states the verb, resource, and scope. Calling it a 'unified mutation tool' distinguishes it from read-only siblings like get_objects and get_view. Although it doesn't enumerate all operations, the enumeration is present in the input schema's operation enum.

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 description provides clear context: it is for mutations on a CAD document, with session behavior explained (recording steps, rollback via session_rollback). It also explicitly directs users to operation_help for per-operation parameter references, which is an important alternative source. However, it does not explicitly state when to use this tool versus other mutation-like siblings (e.g., assemble, execute_code) or provide exclusions.

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