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

Annotations already declare readOnlyHint/idempotentHint/non-destructive and closed-world, so safety is covered. The description adds real behavior beyond that: it derives duplicates from per-source counts plus records_after_deduplication, validates the arithmetic and returns warnings when numbers do not add up, emits both SVG and PDF, and requires no account. What remains undisclosed is what happens when required counts are absent or when warnings fire.

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 trigger and prerequisite, then the capability, then the constraint, then the no-login note. Dense but nearly every clause carries decision-relevant information; the long parenthetical count list is the one place that pushes toward density over scanability.

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 tool with deeply nested inputs and an output schema (so return values need not be described), the description covers purpose, preconditions, derivation logic, validation warnings, output formats, and the must-not-invent constraint. An agent has everything needed to call it correctly without guessing.

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 71%, and the description compensates by mapping real-world PRISMA stages onto parameters: per-source counts, duplicates, screened/excluded, reports sought and assessed, and per-reason exclusions, plus the 'previous version' arm for updated reviews and the 'other methods' arm. It also adds an important semantic rule not in the schema ('leave unknown ones out', 'pass only numbers the user gave you'). Only lang and the precise label-vs-count shape of exclusion reasons go unexplained.

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 (draws) and a specific artifact (PRISMA 2020 flow diagram) rendered as SVG and PDF, and enumerates the exact inputs it consumes (records per database, duplicates, screened, excluded, reports sought/assessed, exclusion reasons, studies included). An agent can tell this apart from revisia_prisma_checklist or revisia_risk_of_bias_figure without opening the 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?

Opens with an explicit trigger condition ('use this when the user needs ... and already has the counts per stage') and adds an explicit exclusion ('does not search databases, screen records or invent counts'). It stops short of naming a specific sibling (e.g. the checklist tool) as the alternative, so routing to the wrong PRISMA artifact is still inferable rather than stated.

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