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

Beyond the annotations, it discloses the permission tier (owner/admin seats), that execution is asynchronous and results must be read back via training_modules with module_id, and the idempotency contract ('a repeat never starts a second run'). It also warns that regenerate replaces existing questions, which is genuinely useful destructive-behavior context the annotations don't convey. This is exactly the kind of context annotations alone cannot provide.

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?

It is dense but front-loaded: purpose first, then permissions, skip/replace semantics, scope limiting, async polling, and retry rules. The parentheticals and semicolons make it slightly run-on, but no sentence is wasted.

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 background mutation with no output schema, it covers the full lifecycle: what starts, who may start it, what is skipped or replaced, how to poll for status, and how retries behave. The only minor gap is that it doesn't say what the call returns immediately (e.g., an acknowledgment) or how failures surface.

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 only 25%, but the description compensates by explaining regenerate (replaces existing questions) and sop_ids (limits which atoms) in behavioral terms; idempotency_key is already documented in the schema. module_id is left implicit, which is acceptable since it is the obvious subject of the operation.

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 uses a concrete verb+resource ('Starts Sergio writing quiz questions for a module's atoms') and anchors it to the known product action 'Generate questions', so the agent knows this is bulk AI generation rather than a manual edit. It does not name a sibling such as add_training_question to draw the boundary explicitly, which keeps it from a 5.

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 the prerequisite ('owner or admin seats only'), explains that atoms with existing questions are skipped and that regenerate: true replaces them, and that sop_ids narrows scope — enough to choose the right option set. It never names an alternative tool (e.g., adding a single question manually) or says when not to use it.

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