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

Annotations already declare readOnly, idempotent, closedWorld and non-destructive, so the safety profile is covered. The description adds the useful fact that this is invoked automatically by the card UI rather than by an agent, but says nothing about what recomputation produces, whether the spec argument overrides stored state, or how specHash is used.

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 short sentences, front-loaded with the trigger and immediately followed by the routing instruction. Nothing superfluous.

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?

As a tool the agent is told not to call, withholding detail is partly defensible, but the schema still exposes three undocumented parameters and there is no output schema. An agent that does need to invoke it (or reason about what render_model does underneath) gets no information about the contract.

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% on three parameters, including two nested objects, so the description carries the full burden and largely fails. Only the word 'input' faintly ties the inputs object to card input values; specHash and the optional spec are entirely unexplained.

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 identifies the trigger (a user edit on the Valem card) and effectively states what the tool is: an internal recalculation callback. The verb 'recompute' comes mostly from the title, but the triggering-context sentence and the pointer to render_model make its role distinguishable from the sibling tools.

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 states an explicit exclusion ('Not for direct use') and names the alternative ('call render_model instead'). An agent has an unambiguous directive about what to call when it wants card output.

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