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

With no annotations, the description must carry the full behavioral load; it does disclose the completion state and the exact shape of the returned evidence (verbatim text, [ev:id], date, neighbour turns, rationale) plus a 'what is missing' signal, which is valuable. It is silent on what happens to an unknown or expired job_id, retention/cleanup of the job, and the pagination implied by the page parameter.

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?

Three short sentences with the core action front-loaded, followed by the return contract and a usage caveat. No filler or repetition.

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?

Returned content is well specified despite there being no output schema, and the not-an-answer caveat is important. But for a polling tool with no annotations and two undocumented parameters, the definition should also cover job_id provenance and the meaning of 'page' – those gaps leave an agent guessing.

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%, so the description must explain both parameters; it only gestures at job_id ('by job_id') and never says where that id originates. The 'page' parameter is not mentioned at all, leaving its paging semantics invisible.

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 gives a specific verb and resource ('Poll a deep search job by job_id') and explicitly frames the return as evidence rather than a finished answer. It does not differentiate itself by name from siblings such as memory_recall_later or memory_start, but the polling-by-job_id scope makes the intent unambiguous.

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?

'Poll' plus 'When done' implies a pending/complete lifecycle and therefore repeated calls, and 'Not an answer: answer yourself from it' tells the agent what to do with the result. However it never states which sibling starts the job (memory_start / memory_recall_later), whether polling is expected repeatedly, or when to stop polling – the routing context is only implied.

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