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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so safety is covered. The description still adds real value beyond them: transient processing with no retention, a 'use synthetic or anonymised inputs only' constraint, deterministic version-pinned resolution, AP2 artifact export with execution_hash, and browser-delegation behaviour for gpu:true nodes. It does not describe the return payload's shape beyond a field list, and the boilerplate is repeated twice.

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 tool's actual function appears only in the trailing 'Summary:' after a long run of chain-provenance, compute-binding and FV-status boilerplate, with a duplicated 'OpenChainGraph compute node' clause. Front-loading is poor: an agent must read to the end to learn what the tool does.

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 deterministic lookup with no output schema, the description supplies the input dimensions, the determinism/table-versioning contract, the six output fields, and the upstream/downstream chaining context. What is missing is only minor: no pagination/error behaviour, and no explicit statement of what happens for unsupported country pairs.

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 baseline is 3. The description restates the decision inputs (supplier/buyer country, transaction type, transaction date) and repeats the compute-mode enum semantics already in the schema. It adds no format hints for country codes or transaction_date beyond what the schema provides.

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 'Summary:' sentence gives a specific verb and resource: a deterministic lookup over a version-pinned mandate table that routes supplier/buyer country, transaction type and transaction date to an e-invoicing regime, format, phase status and transmission channel. That is enough to distinguish it from data-validation siblings. However, the statement of purpose is buried behind several sentences of compute-node boilerplate, and no sibling tool is named explicitly (only upstream/downstream artifact IDs).

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 guidance on which compute mode to pick (auto/server/browser, gpu:true delegation), but nothing on when to choose this router over the many e-invoicing siblings such as validate_einvoice_format, validate_einvoice_batch, verify_einvoice_vat_calc, or build_einvoice_transmission_receipt. The 'Consumes upstream from / Output feeds' lines hint at pipeline position but do not state exclusions or alternatives.

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