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

Beyond the readOnly/idempotent annotations, it discloses real behavioral traits: hours arrive in REST GLS format (ms + 1h), pauses and closures are presented as supplied, an absent day does not mean closed, no displayed hour proves current opening, no distance or capacity is invented, and no reservation occurs. It also documents that errors carry category/fields/next_action and that a GLS failure should trigger an identical retry without touching addresses.

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 definition opens with six sentences of address-naming prohibitions and a scripted customer phrase before stating the actual search, and it reproduces the schema's destination text verbatim. Workflow, consent and error-retry instructions are genuinely useful, but they are padded and the lede is buried rather than front-loaded.

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?

With an output schema present, return values need not be explained, yet the description still characterizes them (identifier, name, address, readable hours, pauses/closures) and sets expectations that fields may be absent. The one real hole is `lines`, entirely undefined in both prose and schema descriptions; otherwise the tool's role in the quote→order flow is covered.

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?

The long destination paragraph is a near-verbatim duplicate of the schema's own destination description, so it adds no meaning beyond the structured field. The second required parameter, `lines`, is never mentioned in the description at all — an agent gets no hint that these are cart lines used to constrain the relay search, even though schema coverage on it is zero.

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 description names a specific verb and resource — 'Rechercher les points relais GLS autour de destination … via l'API GLS existante' — so an agent knows it returns candidate pickup points, not a selection. It is clearly distinct from quote_cart_shipping, whose relay_id it consumes. However the core purpose statement is buried after several sentences about what the address is *not*, weakening front-loading of the action.

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 explicitly ties this tool into the ordering workflow: take a relay_id from delivery={mode:gls_pickup,relay_id} returned by quote_cart_shipping/quote_cart/create_order, note that the server re-verifies the choice, keep shipping_address vs billing_address separate, and re-obtain consent if relay/address/amount changes. It also gives a retry rule for GLS outages. What it never states is the inverse — when *not* to call it (e.g. when no cart lines exist yet) — so it stops short of explicit exclusion criteria.

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