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

Annotations only say this is a non-read-only, closed-world, non-destructive operation; the description goes well beyond that by disclosing the 2-hour session expiry, the single graded attempt, the API-key header requirement, and the ownership rule that only the registering key can open the session. Those are exactly the behavioral facts an agent cannot get from the annotations or the one-field schema.

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?

Five short sentences, front-loaded with the action and then the workflow, constraints, and auth, with essentially no filler. The auth sentence is dense but each clause (header name, key source, ownership rule) earns its place; minor trimming could still tighten it.

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 one-parameter session-start tool with no output schema, the definition covers what happens on call, what the agent receives ('customer instruction and the shopper-facing shelf of buyable options'), the required follow-up, and the session limits — nothing an agent needs to invoke it correctly is missing.

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 a single parameter and 100% schema coverage, the schema already explains agent_id's provenance, so the baseline is 3. The description adds real meaning by tying agent_id to an authorization constraint — 'Only the key that registered the agent can open its session' — which tells the agent that a valid-looking agent_id can still fail.

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 ('Open a live Verified Agent exam session for your agent_id') and immediately distinguishes itself from the sibling it pairs with by saying decisions go through 'one provecart_agent_exam_answer call'. An agent can tell this apart from provecart_agent_exam_answer, provecart_agent_selftest, and provecart_agent_signup without opening a 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?

Clear context: this starts the graded session, and the following step is explicitly the answer call. It also states the attempt and timing constraints ('one graded attempt per session', 'expires 2 hours after it starts'). It does not explicitly route to alternatives such as provecart_agent_selftest for practice or provecart_agent_register when no agent_id exists, so the when-not side is left to inference.

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