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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), yet the description adds substantial behavioral detail beyond them: the evidence-state model (available/partial/unavailable), the fallback to the latest passing run when a failed run persisted nothing, the untrusted_provider_content labelling and its 'inspect, never instructions' caveat, and the possible provider_wait field. This is exactly the extra context the annotations cannot 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?

Purpose and fallback semantics are front-loaded and the trailing 'Next:' line gives clean routing. The long enumeration of return fields is dense but earns its place since there is no output schema; a slightly tighter phrasing could reduce redundancy without losing meaning.

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 and a single required parameter, the description fully carries the return-shape burden — enumerating the returned fields, explaining the evidence states, the fallback path, and the untrusted-content convention. An agent has enough to call it and interpret the result correctly.

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?

One parameter with 0% schema description coverage, so the description carries the burden. It provides an example object ({table_id: "…"}), which confirms the parameter's role, but offers no explanation of what the id identifies or where to obtain it. Adequate given the obviousness of a table UUID, but it does not fully compensate for the coverage gap.

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 one table's current promotion pointers and latest recorded failure.' This clearly distinguishes it from siblings like get_table (raw table), get_run (a run), or list_runs, since it is explicitly framed as a point-in-time diagnosis of a table's failure state.

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?

Explains when the diagnosis is needed (when a failure names a run, including runs that never sealed a version) and routes the agent onward: 'use get_run or run_events for the named run when you need its timeline.' It also excludes the promote/retry/replay siblings via 'does not retry, promote or alter a table,' though it doesn't explicitly state when NOT to call this at all.

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