audit_column

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

Beyond the readOnlyHint=true annotation, the description fully explains the behavior: per-row recalculation using x=varColumn value and i=row sequence, a default threshold of 1%, and reporting only deviating rows. It also discloses a dependency on list_tables column patterns. No contradiction with annotations.

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 sentences with no wasted words: the first front-loads the purpose and algorithm, and the second appends a short actionable prerequisite. Every sentence earns its place.

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 5-parameter tool with no output schema, the description covers core semantics, default threshold, use case, and a prerequisite. It does not specify the exact shape of the returned rows (e.g., whether each row includes row index, original value, or deviation), but the combination of 'rows' and formula-variable explanation gives a reasonably complete picture.

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 the baseline is 3. The description adds meaningful value by defining expression semantics (x and i, with an example), noting varColumn is typically a rank/sequence column, and specifying thresholdPct's default of 1%. This goes beyond the schema descriptions.

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?

The description clearly identifies a column audit tool: it recomputes each row of a specified column against an anchor formula and reports rows that deviate beyond a threshold, with a specific goal of finding manual adjustment breakpoints. However, it does not explicitly contrast itself with similar siblings like recon_diff or eval_formula, so it lacks explicit sibling differentiation.

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 gives a concrete prerequisite ('先看 list_tables 的 columnPatterns 了解列的既有模式') and an intended use case (identifying hand-tuned breakpoints in growth curves). It stops short of naming alternative tools or giving when-not-to-use guidance, so it is below the 5 level.

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