grafica_cientifica

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

With no annotations, the description must carry the behavioral burden. It does disclose real side effects: creating a new diapositiva_figura card with an origin footer, versus drawing on top of an existing slide with theme text colors and a transparent background. However, it never says the document must be open/selected, whether the operation is reversible or how to undo it, nor what the call returns (a slide index? an object id?), despite being a document mutation.

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 core identity and the capability list come first, followed by the two modes and finally the sibling alternative — a sensible front-loaded order with no filler sentences. The capability enumeration is dense but every item corresponds to a real parameter, so little is wasted; it could still be tightened into grouped clauses.

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 22-parameter, deeply nested tool with no output schema and 45% schema coverage, the description covers the drawing modes and feature set but leaves the agent without the two required inputs' role (documento must reference an open presentation, series must be non-empty), the positioning semantics, or any return-value expectation. Adequate for a first call, incomplete for confident invocation.

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 only 45% across 22 parameters, so the description needs to compensate. It does map several features to parameters implicitly (error bars symmetric/asymmetric, 'cotas superiores' → limite_superior, escalas → escala, 'bandas de confianza' → banda/y_inf/y_sup, histogramas → valores/bins, segundo eje → eje_y2, lineas_ref, anotaciones), but explains nothing about x/y placement, alto/ancho, leyenda, rejilla, despues_de or tamano_texto defaults. Useful but partial.

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 (scientific chart rendered via matplotlib into vector PDF with theme palette) and enumerates concrete capabilities: error bars, upper-limit arrows, log/symlog scales, inverted axes, confidence bands, histograms, second y axis, reference lines, annotations. It explicitly separates itself from two siblings: 'grafica' for what Keynote draws natively (bars, pie, editable) and 'ecuacion' for the Computer Modern formula rendering used in $…$.

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 explicit alternative and the condition that selects it: 'Para lo que Keynote sí dibuja bien (barras, pastel) usa grafica, que es editable.' It also defines the two operating modes with their conditions — without 'diapositiva' it creates a new diapositiva_figura; with 'diapositiva' it draws over the existing slide in the content area.

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