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

Annotations already declare readOnlyHint and openWorldHint, so the safety profile is covered. The description adds that responses carry an accuracy_tier that varies by engine, which is useful context, but it says nothing about caching, latency, or whether capability data is stable across calls.

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?

Two short sentences, front-loaded with the action and followed by the invocation order. Nothing is redundant or padded.

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 the description need not explain return values, and it correctly covers purpose and call ordering. It could go slightly further by indicating what the accuracy_tier levels imply for downstream trust, but nothing essential is missing.

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%, but the single parameter is an enum (bigquery/snowflake/databricks) whose meaning is self-evident, and the description identifies it as 'the given warehouse engine'. It adds no format or constraint detail beyond that, so the baseline 3 is appropriate for a trivially self-documenting enum.

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 precise verb and resource: it reports, per warehouse engine, which cost signals are exact versus approximate. That purpose is clearly distinct from the cost-estimation and bounded-execution siblings, so an agent can tell what this tool produces without opening the schema.

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?

Explicitly prescribes sequencing ('Call this before estimate_query_cost or run_query_bounded') and names both alternatives by name, plus the reason to call it (interpreting accuracy_tier). This is the strongest form of when-to-use guidance.

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

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

Annotations cover safety (readOnlyHint, openWorldHint), and the description adds genuine behavioral value by decoding the response's accuracy_tier vocabulary (PRECISE/UPPER_BOUND/HEURISTIC) so the agent knows how much to trust the result. It does not disclose latency, rate limits, or failure modes, so it stops short of 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?

Three tight sentences, front-loaded with the core action, then the usage rule, then the interpretation key. Every sentence carries distinct information with no filler.

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 read-only estimation tool with an output schema present, the description supplies the two things an agent most needs: when to call it and how to interpret accuracy_tier. The only gap is guidance on the optional warehouse argument, which is minor given the overall coverage.

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% and there are three parameters, yet the description says nothing about sql, engine, or warehouse semantics. The engine enum values are self-explanatory, but the optional warehouse parameter and what it selects are unexplained in both schema and description.

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+resource ('Estimate the cost of a SQL query') with a clear temporal scope ('before running it'). An agent can distinguish it from run_query_bounded (which executes) and describe_engine_capabilities (which describes) without opening a schema.

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?

Explicitly prescribes when to call it ('ALWAYS call this before running an expensive-looking query'), giving a clear precondition. It does not name the sibling run_query_bounded as the follow-up action, so the routing is implied rather than spelled out.

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

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

Annotations only declare readOnlyHint=false and openWorldHint=true; the description goes well beyond them by disclosing the cost-bound gate, the 'refused' status sentinel, the result.hint diagnostic, and the real monetary/quota side effect of a successful call. This is exactly the extra context a mutating, cost-bearing tool needs.

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?

Front-loaded with the core behavior and free of padding; the parenthetical about refusal and hint is the only slightly dense clause but it carries real information. Efficient overall.

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 values needn't be documented, yet the description still usefully explains the 'refused'/'hint' shape. Behavior and cost implications are complete for a bounded-execution tool; the only real omission is per-parameter meaning, which keeps this from a 5.

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% across 6 parameters, and the description never explains any of them — engine, warehouse, max_rows, max_bytes_billed, max_estimated_cost_usd are all left to the raw titles. The phrase 'your given bounds' hints at the max_* parameters but gives no mapping, units, or default behavior, so it fails to 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 precise verb+resource ('run a query') plus the governing condition ('only if its pre-flight cost estimate is within your given bounds') and the failure mode ('refuses otherwise'). It explicitly names the sibling estimate_query_cost, so an agent can distinguish the two without opening either schema.

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 the selection condition against the alternative ('unlike estimate_query_cost...'), the refusal semantics, and an explicit anti-pattern warning ('don't call this repeatedly without inspecting the result of each call'). When-to-use and when-not are both covered.

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