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

Annotations already declare readOnlyHint=false, idempotentHint=true and destructiveHint=false, so the safety profile is covered. The description adds real value beyond that: no paid generation is incurred, original bytes are imported server-side in one operation, request_id should be reused on uncertain retry, ownership is required, and inputs expire in 24h. It does not say what the call returns (e.g. the created node_id), which limits it to a 3.

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?

Dense and front-loaded: the trigger condition comes first, the sibling disambiguation second, then branch mechanics. Every sentence contributes a routing rule, a parameter branch, or a retry/expiry constraint; none is filler.

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 7-parameter mutation with a nested file object, no output schema and no annotation help on returns, the description covers the decision-critical parts (file vs reference, node_id creation vs fill, retry, expiry, ownership). The notable gap is return shape and how a caller learns the resulting node id, plus the meaning of file_media_family.

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?

With 0% schema description coverage the description carries the load, and it explains the file-to-host binding via openai/fileParams, the node_id branch (specify for an empty existing media node, omit to create a new one), request_id retry semantics, and reference as the no-host-file alternative. It leaves file_media_family and the expected_rev concurrency meaning unexplained, so not 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?

States a specific verb and resource ('adds an attached FILE directly into an owned Canvas project as a media node') and explicitly names the sibling it is not (aimixer_canvas_attach_result for finished generation results). An agent can route between this, attach_result, and media_import without opening any 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?

Gives explicit when-to-use ('user attached a photo/video/audio FILE'), when-not ('for an already finished generation result use aimixer_canvas_attach_result'), and a fallback path ('with no host file use an existing owned reference from aimixer_media_import'). Also states the prerequisite ordering: read Canvas for project_id and expected_rev first.

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