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

Even though annotations declare readOnlyHint=false and openWorldHint=true, the description adds meaningful behavioral context: it starts a run, waits, and returns a result. It explains the idempotency_key behavior (retrying returns the original run) and the timeout condition ('wait_timed_out' then call wait_for_run). The description also mentions approval_mode options, which are not fully covered by the annotations. No contradiction with annotations.

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?

The description is two sentences, front-loaded with the main purpose and outcome. It packs essential information (what it does, capabilities, timeout behavior, and alternatives) without any wasted words. The structure is highly readable and directly actionable.

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?

Given the tool's complexity (an agent delegation with multiple parameters) and the presence of an output schema, the description covers all necessary aspects: it explains what the agent can do, how long it waits, what happens on timeout, and how to recover. The existence of an output schema means the return structure is documented, and the description adds the key detail that `result` contains the final reply. Nothing essential 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?

Schema coverage is 100%, so the baseline is 3. The description adds value by explaining timeout_seconds default/max and the exact action to take on timeout (call wait_for_run with the run id). It also implicitly clarifies that `message` is a natural-language directive and that idempotency_key prevents duplicate runs. These are not in the schema descriptions, pushing the score above the baseline.

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 specific verb ('Delegate a job') and names the resource ('the Crevio agent') and the outcome ('returns the run with the agent's final reply in `result`'). It explicitly distinguishes itself from sibling tools by stating 'For anything longer, use start_task and wait_for_run.' The mention of capabilities (products, customers, orders, email, socials, sites, research) further clarifies scope.

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?

The description provides explicit usage context: it says to use this for short tasks that wait up to timeout_seconds, and for longer tasks to use start_task and wait_for_run. It also describes the timeout behavior and instructs calling wait_for_run when wait_timed_out occurs. This gives clear when-to-use vs when-not guidance with specific alternatives.

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