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

Annotations already declare the safe, idempotent read profile, so the bar is lower, and the description adds genuinely useful non-annotation context: inputs are processed transiently and not stored/logged/retained, the run exports an AP2 artifact carrying an execution_hash, and it chains from an upstream artifact. These are real behavioral disclosures beyond the structured fields. It does not describe the return shape or failure modes, which keeps it out of 5 territory.

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 text is a dense block of repeated plumbing: compute-mode behavior duplicates the schema's own parameter description, and the FV-status hash URL plus the 'snapshot, not a subscription / verifies offline' aside are noise an agent does not need to invoke the tool. Front-loading with the title verb is the one structural virtue, but several sentences do not earn their place.

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 whose actual work lives in the opaque policy_parameters object and which has no output schema, the description omits exactly what an agent needs: which fields to supply to run a verification and what the result contains. It redirects to describe_tool and 'the tool's manifest,' leaving the core invocation semantics incomplete despite the rich infrastructure detail.

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 description coverage is 100%, so the schema already documents compute, parent_hashes, parent_tool_ids, and policy_parameters — baseline 3. The description adds one useful mapping (the upstream parent is art-116), which lightly clarifies parent_hashes/parent_tool_ids, but it explicitly defers the decision-function fields ('See the tool's manifest') and restates compute-mode semantics already in the schema.

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 name and title convey the specific verb+resource (verify a luxury-good product's authenticity), but the description body never describes the verification decision itself — it is almost entirely infrastructure boilerplate about compute binding and chain provenance. The closest thing to differentiation is the note that it consumes an upstream artifact from art-116-product-lineage-builder, but closely related siblings (build_product_lineage, assess_suspect_product_status, resolve_recall_trace) are neither named nor contrasted. An agent learns the topic but not what the tool actually decides.

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?

There is no when-to-use guidance or routing against alternatives. The only usage directive is 'Use synthetic or anonymised inputs only,' which is a data constraint rather than a selection criterion. An agent cannot tell from this text when to pick this tool over the sibling lineage/status/traceability tools.

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