Send a conversion event

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

With no annotations, the description carries the full burden and does so well: it discloses the processing pipeline (normalize, SHA-256 hash, dedupe) and, critically, that Meta sends live while Google is validate-only and LinkedIn is a dry-run — side-effect facts an agent would otherwise get badly wrong. It also states billing impact and dedupe economics, leaving only rate limits and auth mechanics unspecified.

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 core action is front-loaded in the first clause, followed by side-effect disclosure, retry rule, and cost in descending order of importance. Dense but every sentence carries distinct information; the parenthetical event list is the only mild redundancy with the event_name schema examples.

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 9-parameter mutation with a nested user object and no output schema, it covers the highest-risk unknowns: per-destination behavior, dedupe, and billing. The remaining gap is what a successful call returns (accepted/rejected per destination) and how per-destination failures are surfaced, which an agent would want before treating the call as done.

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 67%, and the description adds real meaning on top of it by explaining what the destinations enum values actually do per platform and why event_id must be stable across retries. The value/currency pairing and timestamp default are left to the schema, which covers them adequately.

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 opening clause states a specific verb and resource ('Send one server-side conversion') and enumerates the event kinds it accepts. It is unmistakably distinct from the siblings get_stats and test_meta_connection, which are read/verification tools rather than event ingestion.

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?

It gives explicit operational guidance for retries ('Reuse the same event_id on retries; duplicates are free and not re-sent') and surfaces the cost model that selects for careful batching. It does not, however, tell the agent when to set test=true versus calling the sibling test_meta_connection to validate credentials first.

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