infer_table_relation

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

Annotations provide readOnlyHint=true so the agent knows this is a read-only operation. The description adds behavior such as '确定性拟合', tolerance handling (取整容限内免误报), and default behavior (auto-pairwise on all numeric columns) which goes beyond the schema. However, it doesn't elaborate on output format or failure behavior, but annotations cover the safety profile.

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 description is dense and packed with domain-specific terminology in Chinese, but it front-loads the core function first. It includes example structures and sibling references but could be slightly more concise. It earns a 3 because while informative, it's somewhat dense and may bury the simplest summary.

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?

Given the tool's moderate complexity, the description explains the two modes (pair-specific vs auto-pairwise), default behavior, tolerance semantics, and relationship to sibling tools. There's no output schema, so the description carries the burden of conveying behavior, and it does so adequately. Minor gaps: no explicit statement of what 'deviation rows' output looks like, but context is otherwise complete.

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 schema documents all three parameters. The description adds context: columns default to all numeric columns, thresholdPct tolerance semantics (relative tolerance, max with rounding tolerance). This is useful but the baseline with full schema coverage is 3.

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 states the tool's purpose: infer linear column relationships (B ≈ A × k) and identify deviating rows, or automatically discover baseline anchor columns and coefficients across numeric columns. It distinguishes this from infer_column_rule explicitly, making its scope unambiguous.

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 explicitly states when to use it: for reconstructing generation structures like 'equipment attributes = baseline × equipment coefficient', and mentions comparing results with coefficient tables like read_table. It explicitly excludes column-internal temporal patterns, directing to infer_column_rule.

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