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

Annotations already declare read-only/idempotent/no-destruction, but the description adds substantive domain behavior: amounts are gross base pay for full-time work, family supplements, allowances and overtime are excluded, and the figures are reproduced from published tables set by the agreement parties or by law. It also states the default-year behavior and the discovery-then-refetch pattern for occupations.

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?

Three short paragraphs that front-load what the tool does and then move to invocation and output caveats; nearly every sentence carries new information. Minor redundancy between the opening and the second paragraph's restatement of agreement/grade/step, but not enough to hurt scanning.

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 lookup tool with five optional parameters and an output schema, this covers everything an agent needs: what to pass, what comes back (monthly gross, annual figure, validity dates, full step ladder, step intervals), what is deliberately excluded from the figures, and where to go for net pay. Nothing material is left to inference.

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; the description goes beyond it by explaining the interaction between parameters that the schema does not encode — agreement+grade together, step optional to get the full grade ladder, or occupation as a substitute for both. It stops short of explaining step syntax or year-range semantics that the schema already covers.

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?

Names a specific verb and resource ('look up German public-sector pay') and enumerates the exact table families covered: collective agreements for public, church and welfare employers, plus statutory Besoldung for the federation and each state. It is clearly separable from siblings such as get_salary_benchmark and calculate_net_salary, which the description itself routes to.

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 explicit triggers ('when the user names a pay table with a grade... or asks which pay table applies to their occupation') and the conditional branch: pass agreement+grade directly, or pass occupation first and call again. It also names the alternative for a different need ('For take-home pay call calculate_net_salary... set employment to civil_servant').

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