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

The description adds substantial behavior beyond annotations: snapshot storage, immediate checking against BLS/Open Food Facts, correction only when a source backs it, escalation to a human moderator from the second report, and explicitly that no food is deleted and the user's log entry is untouched. That said, the annotations declare destructiveHint=true and idempotentHint=false while the text emphasizes that nothing is destroyed, and the description never addresses what happens on repeat submissions or what the call returns, so the tension is softened rather than fully resolved. Strong added context, but not a complete behavioral account for a flagged-destructive mutation.

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?

It is front-loaded with the purpose and defect kinds, then layers consequences and safety constraints, with the most important negative guarantees stated last for emphasis. The app-analogy sentence ("Same path as the report button") earns its place as a mental model, though the middle section is a dense run of clauses that could be split for readability.

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 6-parameter tool with a nested evidence object and no output schema, the description covers the lifecycle (snapshot, immediate source check, conditional correction, moderator escalation) and the key non-effects (no deletion, no writes from suggested values, log entry unchanged). The main remaining gap is what the call returns to the caller, which an agent may need for confirming the report succeeded.

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 83%, so the schema already documents the parameters well; the description's contribution is mainly behavioral, clarifying that foodItemId refers to the catalog id and that suggestedPer100g and evidence are recorded and compared against reference sources rather than written. The schema description for suggestedPer100g already says "Recorded, never written," so much of this is restatement. 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 first sentence gives a specific verb and resource ("Report that a food in Joiva's shared catalog is wrong") and enumerates the four defect kinds, which map directly to the kind enum. It is clearly distinguishable from siblings like joiva_update_log_entry or joiva_log_food, and the closing note that the user's log entry is unaffected separates it from logging tools.

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 states the triggering condition twice with concrete examples ("fat can't be right, it was porridge with water" and "clearly implausible"), so an agent knows when to reach for it. It does not explicitly name an alternative tool or say when NOT to report (e.g. for a user's own log entry), which keeps it below a 5.

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