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

Beyond the readOnly/idempotent annotations, the description adds real behavioral context: results are immutable observations, no retry occurs, pages are bounded, and a page binds to an exact result revision. The caveat that undelivered fields are not read-proven is a meaningful honesty signal. It stops short of stating what happens on a revision mismatch, but the bar is lower given full annotation coverage.

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 tight sentences with the primary action front-loaded, followed by exclusions and then pagination semantics. Every clause carries information rather than restating the name. Slightly dense phrasing ('grant read proof') costs a little immediate comprehension but not enough to waste space.

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?

An output schema exists, so return shape needn't be described, and annotations cover the safety profile. What remains thin is the input story for a four-parameter tool with zero schema descriptions: an agent could read this and still not know how to source call_id or work_id, or what a failed expected_revision means. Adequate for the read semantics, incomplete on invocation.

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 carry all four parameters. It gestures at cursor ('bounded pages') and expected_revision ('binds to an exact result revision'), but never explains how work_id differs from call_id or how either is obtained. The 'selected native session' phrasing hints at scope without defining it. Partial compensation only.

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: reads the immutable observation for a call, or its native job status. The reference to 'a call' implicitly ties it to call_studio_tool, but no sibling is named explicitly, so differentiation relies on inference. The abstraction ('observation', 'read proof') is unusual but consistent with the domain vocabulary used across the native.* family.

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 description implies usage ('read the observation for a call') and gives one negative constraint ('does not retry work'), which steers agents away from expecting a re-execution. However it never says when to prefer this over call_studio_tool, native.get_run_status, or native.read_work, all of which sit in the sibling list and plausibly overlap. Guidance is implied rather than stated.

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