retry_download

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

With no annotations, the description carries the full disclosure burden, and it does well: it states that the original download kind, concurrency, export format, report, and task ID are preserved, that current config/account are used, that images come from upstream cache, that completed items are not re-downloaded, and exactly which conditions yield an error. It omits permission/auth requirements and whether the call is side-effecting on the original task, which keeps it below 5.

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 three tight lines that lead with the action and then add semantics and error conditions; every sentence carries information. The only inefficiency is that the cache/config details could be compressed.

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?

Because an output schema exists, return values needn't be explained, and the description covers the mutation's effects, state preconditions, and failure modes well. The main gap is that it doesn't say what to do next when the error is returned, for example which tool to call or how to inspect task state.

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?

There is a single parameter, task_id, and the schema provides no description at all (0% coverage). The description implies a key semantic constraint — the ID must refer to a terminal (non-running) task with unfinished items — which is genuinely useful, but it doesn't spell out what task_id is or where to get it, for example from get_download_task.

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 states a specific verb+resource: create a new background task for the unfinished items of a terminal task, preserving the original report and task ID. This implicitly distinguishes it from start_download/download_batch by scoping it to an already-terminal task, but it never names those siblings explicitly.

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 clearly describes the precondition (the original task must be terminal) and two negative cases that produce an error: no unfinished items, or the task still running. It gives clear context for when to call it, though it doesn't explicitly route the agent to an alternative tool for those error cases.

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