Evaluate formula

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

Annotations only declare readOnlyHint and openWorldHint; the description goes well beyond them. It discloses the return shape and the four possible `type` values, the critical distinction that Excel errors come back as a normal result while parse/unsupported-function failures are tool errors, the supported-function surface (278 functions, enumerated by category), and the odd placement rule that the formula lands in row 1 one column right of the data.

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 action and an example call, then progressively discloses type rules, return shape, and error semantics. It is dense but nearly every sentence carries real information; the long function-category list is the one portion that could be trimmed without much loss.

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?

There is no output schema, and the description fully compensates by specifying the {formula, result, type, stats} response and the meaning of each type value. For a tool with two well-defined params and a nontrivial evaluation model, it covers the prerequisite (API key), the edge cases (errors vs. tool errors, array expressions), and the scope boundary against siblings.

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, but the description earns more by explaining how values are interpreted: true/false are logical cells that SUM/AVERAGE/COUNT skip, an exact error code string becomes that error, and everything else is text. It also adds the non-obvious constraint that the formula is placed one column right of the data, so that cell must not be referenced — a rule absent from the schema.

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 and resource ("Evaluate an Excel formula and return the computed result") and immediately bounds its scope as a quick pre-flight check rather than workbook-wide evaluation. It explicitly names the sibling that handles the other case ("To check formulas that already live in a workbook, use create_session instead"), so an agent can route 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 an actionable when-to-use ("check what a formula returns on a few sample values before you put it in a workbook") and an explicit when-not-to-use with the alternative named (create_session for formulas already in a workbook). It also calls out the prerequisite that the tool runs on the Basis API and needs an API key.

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