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

Annotations only declare a non-read-only, non-destructive, non-idempotent write; the description adds the real behavioral contract: chunk protocol, that final=true must ride the last chunk, that identity fields may arrive with any chunk, and that the trust report is returned once finalized. It stops short of saying what happens on rejected/duplicate chunks or invalid review rows, which matters given idempotentHint=false.

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?

Front-loaded with the one-shot case first, then the escalation path, then the exclusivity caveat, then the return value — a sensible order for an agent deciding which route to take. The justification clause about retyping is a slightly long aside, but it earns its place by reinforcing the usage constraint.

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 19-parameter, zero-required, no-output-schema ingestion tool, the description covers mode selection, batching limits, correlation keys, and the return value ('the trust report when finalized'). Combined with 68% schema coverage, an agent has enough to call it correctly; remaining gaps are field-level validation details rather than call-shaping information.

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 68%, so many fields are documented structurally, but the description still adds meaning the schema cannot: the role of submission_id as the chunk-correlation key, and the fact that identity fields are chunk-order-independent. It says nothing about how lat/lon/name/address relate to placeId/cid resolution, leaving that inference to the agent.

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 ('Submit the review feed inside the tool call') and immediately splits it into two modes: a one-shot whole-payload submission and a chunked flow that depends on begin_submission. That split lets an agent tell this tool apart from begin_submission/finalize_submission without opening a single 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?

Explicit when-to-use ('only when you cannot POST to the upload_url yourself (no page fetch, no curl)'), explicit alternative, and explicit thresholds ('fine up to about 150 reviews', chunks of up to 150). The exclusion is backed with a reason (retyping is slow and error-prone), so the agent knows why the other route is preferred.

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