BringMeThat

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

Goes far beyond the annotations: it discloses that no card is charged and no order is placed, that failures return the problems to fix, the full price composition, premium waivers, and the card authorization buffer (+$15 over $30, otherwise 45%). This is exactly the behavioral context an agent needs before telling a customer to open a link.

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?

Purpose and non-effects are front-loaded, but the pricing and card-authorization paragraphs are dense and largely explanatory rather than needed to invoke the tool correctly. Several sentences earn their place; the premium tiers and 45%/$15 buffer detail are borderline.

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 no output schema, the description does tell the agent what comes back (checkout_url, prices, or problems to fix), which is the key return-value information. It is weaker on the meaning of restaurant_id and the delivery_address prefill behavior, but overall complete enough for correct invocation.

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 only 25%, and the description adds partial meaning for items/options ('validates items and required options first') and fulfillment (delivery fee $0 for pickup). It says nothing about restaurant_id, quantity limits, or delivery_address, so the compensation for the coverage gap is incomplete.

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?

States a specific verb+resource ('Build a bringmethat.com checkout link for the customer's items') and immediately bounds the effect ('Does not charge a card and does not place the order'), which cleanly separates it from get_menu, get_order_status, and search_restaurants.

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?

The description makes the trigger clear (customer is ready to review total and pay) and says it validates items/options before proceeding. It does not explicitly route to siblings, e.g. that item_id and option names must come from get_menu, leaving that workflow inference to the schema.

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

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

Annotations already declare readOnlyHint and idempotentHint, but the description adds real behavioral context the annotations don't: what the response contains, that prices are item-only, and that fees are added at checkout per create_checkout_link. It does not mention result size limits or pagination, which would matter for large menus.

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?

Three compact sentences with no filler; the payload contents come first, then the price/fee caveat, then the parameter guidance. Every sentence carries information an agent would otherwise have to guess.

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 no output schema, the description does the work of explaining what comes back, including the required-option-group detail that affects downstream checkout. Only the absence of size/pagination expectations keeps it from being fully complete for a potentially large menu response.

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 undescribed parameter (restaurant_id) is covered by the description's pointer to search_restaurants. The description also explains the intent of section vs search — narrowing an oversized menu — which goes beyond the schema's per-field examples.

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?

States a specific verb and resource (get a restaurant's menu) and enumerates the returned payload: sections, items with item_id, descriptions, prices, and option groups including required flags. That level of specificity makes it unmistakable against siblings like search_restaurants or get_order_status.

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?

Explicitly routes the agent on where restaurant_id comes from ('Use restaurant_id from search_restaurants') and gives a narrowing strategy for large menus via section or search. It stops short of stating when not to call it or naming an alternative tool for a different need.

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

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

Annotations already declare readOnlyHint/idempotentHint/non-destructive, and the description reinforces this ('Does not change the order') while adding a real behavioral trait not in the structured data: results are gated on both identifiers matching. It stops short of describing what status values come back or any rate limiting.

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?

Three short sentences: what it does, that it is non-mutating, and the matching precondition. Front-loaded with the purpose and no wasted clauses.

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 two-parameter read tool with no output schema, the description covers purpose, safety and the authentication gate. The only gap is that it does not hint at what the returned status looks like, which matters somewhat given there is no output schema.

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 only 50% — the email parameter is undocumented in the schema — and the description compensates by specifying that it must be the email used at checkout, plus that order numbers start with 'R'. That is meaningful added semantics beyond the single schema example.

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?

States a specific verb and resource ('Read the status of a BringMeThat order'), which is clearly distinct from the sibling tools create_checkout_link, get_menu and search_restaurants. An agent can route to it without opening the schema.

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?

The description gives the concrete precondition for use (order number starting with R plus the checkout email, both required to match), which is effectively the usage rule. It does not name alternatives or explicitly exclude scenarios, but the siblings are unrelated operations so no routing ambiguity remains.

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

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 usefully adds that results are US-only and discloses conditional return behavior — distance only appears for street addresses, Yelp rating only when on file — which is real context an agent can't get from annotations.

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?

Three sentences, front-loaded with purpose, then return values, then the pre-call prerequisite. No filler and nothing repeated from the schema verbatim.

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?

No output schema exists, and the description enumerates the returned fields, so an agent knows what to expect. For a 4-parameter read tool with full annotation coverage, the only minor omission is the limit parameter's behavior.

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?

With only 50% schema coverage, the description compensates for the undocumented parameters: it restates the accepted location formats and ties the fulfillment choice (delivery vs pickup) to the result set, and explains why a street address matters (distances). It says nothing about the undocumented 'limit' parameter (default 8, max 10), leaving that one gap.

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?

States a specific verb and resource ('Find restaurants') plus the scope that defines it: US locations, delivery or pickup. That scope cleanly separates it from siblings like get_menu and get_order_status, which are lookup tools for an already-chosen restaurant.

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?

Adds an actionable prerequisite: obtain a ZIP code, city/state, or street address from the user before calling. It doesn't name alternatives or exclusions (e.g., what to do when nothing is found, or that get_menu is the next step), so it stops short of full when/when-not guidance.

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