neurover_cleanup_step

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

The annotations already declare destructiveHint=true, openWorldHint=true, and readOnlyHint=false, so the safety profile is covered by structured data. The description adds real value by stating that only already-approved items are removed and by disclosing a secondary after-failure verification mode. It does not explain irreversibility, partial-failure semantics, required permissions, or whether a second call resumes or repeats the batch, which a destructive pipeline step should cover.

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 a single short sentence with the step identity front-loaded, so it is not bloated. But the trailing "также проверка..." clause is bolted on without structure and creates ambiguity rather than compactness, so the terseness costs more than it saves.

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 destructive, open-world step in a multi-tool cleanup pipeline with no output schema and zero parameter documentation, the description is thin. It omits the meaning of jobId, the batch/wait tradeoff, and what the agent should do after a partial or ambiguous failure, leaving an agent unable to call it correctly in the failure-recovery case it explicitly advertises.

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% across three parameters, so the description carries the full burden and largely fails. "Одна порция" loosely maps to batchSize but gives no guidance on choosing a size within the 1-50 range, and neither waitSeconds nor the required jobId is mentioned at all – an agent cannot tell whether jobId is the cleanup job from cleanup_start or something else.

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 first clause gives a specific verb and resource ("удалить одну порцию одобренного" – delete one batch of approved items), which is more than a restatement of the name. However, "шаг 5 (ручной режим)" references an undefined step pipeline, and the appended clause "также проверка неясного результата после сбоя" makes it unclear whether this is fundamentally a delete or a verify tool. It does not distinguish itself from cleanup_approve, cleanup_verify, or cleanup_start, so an agent cannot reliably route between them.

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?

"Ручной режим" implies an automatic counterpart exists, and "после сбоя" hints at a recovery-use case, but neither is spelled out as a when-to-use rule. There is no statement of when to prefer this over cleanup_start, cleanup_approve_files, or cleanup_verify, and no prerequisites such as needing a prior approval step. Guidance is inferable at best.

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