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

Annotations already carry readOnly/idempotent/non-destructive, and the description layers substantial extra context: timeout/interruption/read_error produce a labelled last-confirmed snapshot with an exact continuation, unavailable content degrades to metadata-only lifecycle, settled means terminal rather than successful, and reads may reconcile expired leases server-side. This is far beyond what the annotations convey.

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 core purpose is front-loaded and every sentence deals with behavior rather than filler. It is dense and jargon-laden across six packed clauses, which costs some readability but is defensible for the amount of behavior being disclosed.

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?

With no output schema, the description rightly explains return semantics (settled vs metadata-only lifecycle, last-confirmed snapshot, continuation), which is the main thing an agent needs. Minor gaps remain on how the two timing parameters relate, but the risky behavioral territory is covered.

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 0%, so the description must carry parameter meaning, and it only partly does: it implies the explicit project/environment identifiers and the 30s cap (matching timeoutMs max 30000), but batchId and pollIntervalMs semantics and their interaction are left unstated. Partial compensation for a 5-parameter, undescribed schema.

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 (observe one existing batch) with scope (explicit project/environment) and a hard bound (at most 30 seconds). The negative framing (starts no target, claims no work, cancels nothing, requests no assessment) cleanly separates it from create/cancel/assess siblings.

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 exclusions (does not start, claim, cancel, or assess) give an agent clear context for when this tool applies versus mutating siblings, and the 30s observation window is implicit guidance. It stops short of naming an explicit alternative or a when-not condition, so it is not a full 5.

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