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

Annotations already declare readOnly/idempotent/non-destructive, but the description adds substantial domain behavior: credit-to-micros conversion, that failed and pre-start-cancelled runs are free, that post-start cancellations pay recorded spend, that credits return at test end, and the directional=true exclusion semantics. This is real value beyond the safety profile, though the charge formula is dense and unverified.

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 first sentence is well front-loaded and names the returned fields, but the body then dumps several billing edge cases (unknown-spend formula, cancellation charging rules) that read more like documentation than selection criteria. It is overstuffed rather than lean.

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 carries the burden of explaining return values, and it does name the estimate fields and per-run table plus the directional flag meaning. Given the tool's complexity, it is largely complete, though it could clarify the shape/structure of the per-run table.

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?

The tool takes zero parameters, so the schema offers nothing to clarify; baseline for a no-param tool is 4. No parameter-level guidance is needed or missing.

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 (read) and resource (latest paid Agent Picks test status) and enumerates the returned quantities (estimate, reserved/settled/refunded micros, per-run table). It partially differentiates from siblings by routing install rates to devtune_get_agent_picks_summary and noting it cannot start a test, though other close siblings (get_agent_picks_run, get_agent_picks_history) go unmentioned.

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?

Usage is implied (read the latest paid test's rollup) and one alternative is named for install rates, plus the negative constraint 'cannot start a test'. But there is no explicit when-to-use vs the many adjacent siblings like get_agent_picks_run or list_agent_picks_runs, leaving selection partly to inference.

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