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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, so the low safety burden is covered. The description adds meaningful behavioral context beyond annotations: it discloses input handling ('uses only what you pass in — a missing problem renders as a visible bracket, never an invented one'), which is a critical faithfulness guarantee. It also discloses provenance ('From 3,400+ mentoring sessions at marian.coach') and localization options (English/Czech, tykani/vykani). It doesn't contradict annotations; it builds on them.

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 long but information-dense; every clause adds a deliverable or a constraint. It front-loads the purpose ('Build the case that gets your company to pay...') before enumerating outputs. It is arguably over-stuffed (a single run-on sentence with many parentheticals), but none of the material is filler, and the structure is a coherent list of the tool's output artifacts.

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 17 params, an output schema, and full schema coverage, the description is remarkably complete. It explains what the tool produces (emails, one-pager, talking points), how it behaves with missing input (visible bracket, no invented content), what personalization knobs exist (role, language, formality, budget situation), and how the user's context flows into the output. The required 'context' parameter is explicitly described in the schema, and the description adds the 'Uses only what you pass in' guardrail, which is important for trust. The only minor gap is that it doesn't spell out the return format, but the presence of an output schema makes that unnecessary.

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 schema already documents all 17 parameters. The description adds contextual meaning by weaving key parameters into the narrative: 1,975 EUR quarter (6 sessions, 5 paid + 1 free) maps to situation=ld_budget, 395 EUR pilot session maps to situation=no_budget, 'senior people at risk x replacement cost' maps to at_risk_attrition, and the role list maps to role. It doesn't restate each schema property, but it enriches the most decision-relevant ones.

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 a specific verb ('build the case that gets your company to pay for leadership mentoring') with a named resource (marian.coach get-your-company-to-pay page) and explicitly lists the deliverables: ROI formula, CFO rule, worked examples, napkin math, email, one-pager, talking points, objections. This clearly distinguishes it from sibling tools like calculate_developer_value or estimate_coaching_cost, which sound narrower in scope.

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 implies usage: it should be used when the user wants to persuade their employer to pay for mentoring, and it describes content tailored to a manager-facing pitch. It doesn't explicitly name which sibling to use instead (e.g., estimate_coaching_cost), but it does state scope boundaries like 'Uses only what you pass in' and mentions the situation enum (learning budget vs no budget), giving enough context for an agent to select it appropriately.

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