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

Annotations only declare readOnly=false, openWorld=true, idempotent=false, destructive=false — they say nothing about latency or async behavior. The description adds exactly that: up to ~4 minutes runtime, a ~52-second cutoff returning 'still running' with a toolCallId, background completion that saves the result, concurrent same-story calls that start nothing, and the warning that cancelling the call 'stops the run and saves nothing'.

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 persistence are front-loaded in the first sentence, and the long async guidance is packed with operational detail rather than filler. It is on the long side (eight sentences) and the 'still running' repeat-call behavior is restated twice, but that repetition carries real information about concurrent calls.

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 still explains what happens to the result (saved as a design the team can preview/refine/attach) and how to retrieve it via get_tool_call_status. The one gap is that it does not describe the shape of a successful response (design id, option list), which an agent would want when no output schema exists.

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 coverage is 100%, so the schema already documents all three parameters — including the 'ORB-526' preferred format, the 1-5 count range and default, and the 'dark mode, sidebar layout' instructions example. The description repeats those same facts without adding new semantics, so baseline 3 applies.

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 ('Generate AI UI-mockup options (HTML/CSS) for a story') plus the persistence effect, so an agent immediately knows this creates new design artifacts. Sibling-name confusion with refine_ui_design, list_designs, or link_design is avoided because the description names the output artifact (a design) and the source (story title + description).

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 concrete usage context: identify the story by ticket key or UUID, optionally pass count 1-5 and free-text instructions, and explicitly routes long-running runs to stride.get_tool_call_status with the exact waitSeconds 40 pattern. It does not explicitly say 'use refine_ui_design instead when modifying an existing design', so the new-vs-refine boundary is only implied by the parenthetical 'for a refine, the same option'.

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