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

Goes well past the annotations: warns that validate_only is ignored by the endpoint and the tool deliberately errors rather than faking a dry run, that a real object would be created, that no delete/archive tool exists in this toolset, and that a mismatched pixel_id returns {'error': 'foreign_object'}. This is exactly the kind of irreversible-write context an agent needs and annotations (readOnlyHint=false, destructiveHint=false) do not supply.

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?

Dense but each line carries information an agent cannot get from the schema — no filler. The main structural weakness is that the most consequential warning (validate_only aborts, no delete path) sits mid-document rather than being front-loaded, and the German prose mixes parameter listing with pitfalls.

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 10-parameter open-world mutation tool with an output schema, the description covers rule semantics, enum values, unit conventions, error cases, and irreversibility. Nothing an agent needs to invoke it correctly is missing.

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 0% and the description compensates fully: it defines the three mutually exclusive rule parameters, their case-sensitivity differences, the eleven custom_event_type values that are absent from the schema, currency units for default_conversion_value (not cents), and workspace's optionality. Only the obvious name/description params are left implicit.

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 ('Custom Conversion anlegen') and immediately characterizes the typical case (a thank-you page as a lead event). It is clearly distinguishable from the read-side sibling meta_list_custom_conversions without opening either 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?

Gives concrete call preconditions: exactly one of url_contains / url_equals / rule, 'im Zweifel url_contains nehmen', and pixel_id must come from meta_list_pixels. It stops short of naming alternative tools for other conversion types (e.g. ads_create_conversion_action or openai_ads_create_conversion_setting) that an agent might otherwise reach for.

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