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

Annotations only declare readOnlyHint=false and idempotentHint=false; the description adds substantive context beyond them: results are persisted with source run, expression, value/condition state, timestamp, and evidence, and a resolutionId is returned. It also signals determinism, explaining why the call is safe to reason about even though it writes records. Auth/rate-limit behavior and error modes (e.g., both fields supplied) are not covered, keeping it below a 5.

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 is front-loaded in the first sentence, followed by behavioral facts, then the disambiguation rule, then the example. Every sentence carries information, though the inline JSON example is a single dense run-on that is harder to parse than a shorter shape sketch would be.

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 deeply nested anyOf schema with no output schema, the description supplies the return contract (resolutionId plus persisted evidence/state) and a canonical worked example, which is enough to construct a call correctly. Missing details on failure modes and on how the run's completeness is validated leave a small gap.

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 100%, so baseline is 3; the description adds real value on top by clarifying the mutual exclusivity of elementId vs resolution, correcting the field name ('resolution, not expression'), and providing a fully worked nested comparison example with reducer=last on each side. It does not explain analysisRunId's provenance beyond what the schema already says.

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 ('resolve') and two precise resources ('one stable semantic element or one structured expression') scoped to 'the full durable dataset from a completed analysis run'. An agent can distinguish this from inspect_analysis_artifact or run_data_analysis purely from the description, including the correct field name ('resolution, not expression').

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?

Gives a hard usage rule ('Pass exactly one of elementId or resolution') and names downstream intent ('Use the returned resolutionId for a later authorization action'), plus a clear prohibition ('Never copy a displayed chart number into a trade payload'). It stops short of explicitly naming when to prefer a sibling tool or which preconditions (e.g., run must be completed) must hold, leaving some inference.

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