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

Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint=false, so the safety profile is covered. The description adds behavioral context by specifying the exact return fields and decision types (agreement, modification, waiver, compliance check) and the query types supported. This goes beyond the annotations and helps the agent understand what the tool will do and what data it will receive.

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 two sentences. The first sentence defines what a PIP is and the tool's purpose; the second lists returned fields and example questions. It is front-loaded with the essential definition, and every sentence adds value without redundancy. No wasted words, and the structure is clean and scannable.

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 complexity (a search/query tool with 4 parameters, all described in the schema), the description is complete. It covers what the tool returns, the types of queries it answers, and gives concrete examples. No output schema exists, but the description enumerates the return fields, which suffices. There is no missing critical information that would prevent correct 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?

The schema already provides descriptions for all four parameters (limit, query, since, decision_type) with 100% coverage, so the schema carries the primary semantics. The description adds marginal value by giving example queries and noting the decision_type filter examples ('waiver', 'agreeing on', 'compliance'), but it does not significantly enhance the parameter meaning beyond what the schema already explains. Baseline 3 is appropriate.

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 states the specific resource (EMA Paediatric Investigation Plans), the action (returns details about them), and the key fields (active substance, decision type, etc.). It also distinguishes itself from sibling EMA tools by focusing on paediatric plans specifically, and answers concrete queries like 'what paediatric studies has the EMA required for X'. This is a clear, non-tautological purpose.

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 explicit context on when to use the tool by providing example questions ('answers...'). It implies it is for paediatric investigation plan queries, which differentiates from other EMA tools (e.g., ema_chmp_opinions for CHMP opinions). However, it does not explicitly name alternatives or state when NOT to use it, leaving some inference to the agent. Still, the example use cases are strong guidance.

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