audit_formulas

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

With no annotations provided, the description carries the full disclosure burden, and it delivers richly: it declares read-only safety, states it 'works while the file is open in Excel' (concurrency behavior), and reveals the truncation mechanism ('capped with exact counts and a truncated flag'). It also exposes a subtle semantic gotcha — missing_cached_values cells 'read as blank to every non-Excel consumer until a recalculation' — which is exactly the kind of non-obvious behavior an agent needs to know.

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 purpose and scope are front-loaded in the first sentence, followed by a functional per-report breakdown that an agent needs to interpret results. At roughly 140 words it is longer than average, but every sentence earns its place — the report details, truncation behavior, usage window, and safety note are all decision-relevant. A slight trim of the per-report prose would make it tighter, but nothing is wasted.

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?

The description is remarkably complete for a complex tool with zero annotations and zero schema descriptions: it covers scope selection, all five report semantics, truncation limits, concurrency, read-only safety, and practical usage timing, while the output schema handles return-value structure. The remaining gap is the ambiguous interaction of sheet and location parameters together, and the absence of explicit parameter-name mapping — minor issues given how much is already covered.

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%, so the description must compensate, and it does: the opening line maps the three scope modes onto the parameter combination — 'a range' (location), 'one sheet (sheet alone)' (sheet), 'whole workbook (no scope)' (neither). This is essential meaning the schema entirely lacks. It loses one point because the description never explicitly names parameters or defines how location and sheet interact when both are supplied.

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 opens with a specific verb and resource: 'Read-only formula intelligence for a range, one sheet, or the whole workbook,' then names the exact deliverable ('the formula list plus five safety reports'). It distinguishes itself from siblings like read_range (value reading), set_formula (writing formulas), and diagnose_workbook (general diagnostics) by pinning down formula-specific auditing across three explicit scope modes.

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 concrete usage timing: 'Use it before and after structural edits, and to judge whether cached values can be trusted.' This is clear contextual guidance for when to invoke the tool. It stops short of a 5 because it never names alternatives or states when not to use it (e.g., comparing against diagnose_workbook or validate as siblings).

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