toast_graphql

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

Annotations already declare readOnly/idempotent/openWorld, but the description adds real behavior the annotations cannot: enforcement of query-only documents, return of raw data *and* errors, dependency on a signed-in browser session, and the error-recovery contract ('validation errors name the fields the schema lacks'). Missing only volume concerns (rate limits, response size, whether the 8000-char cap applies to variables too).

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?

Front-loads what it does and where, then scope restriction, then examples, then error-handling advice — a natural call order. Every sentence carries actionable content, with no restatement of the tool name or filler.

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 steps in to say it returns raw data and errors, and it covers the biggest failure mode (schema validation errors) with a concrete recovery strategy. For an open-ended escape-hatch tool, nothing an agent needs before calling it is absent.

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 100%, so the schema already defines query/variables/operation_name; the description adds meaning beyond it by naming valid operation shapes (nearbyRestaurants, menusV3 with their input fields) and by framing the query param as accepting fragments only. It adds query-construction guidance the schema cannot express.

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 ('run a read-only GraphQL query against Toast's online-ordering gateway'), names the concrete endpoint, and explicitly positions itself against the sibling set ('for exploring fields the dedicated tools don't cover yet'). An agent can distinguish this from toast_find_restaurants/toast_get_menu without opening any 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?

Gives an explicit trigger ('exploring fields the dedicated tools don't cover yet') plus an explicit exclusion ('mutations and subscriptions are refused'), which steers agents toward the typed sibling tools for known use cases. Known-good operation names with their input shapes give a concrete starting point rather than open-ended guessing.

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