Doorfarer

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

Annotations only cover the safety profile (readOnly, idempotent, open-world); the description adds the material behavior: first search from an unknown town takes ~20 seconds, later ones a second or two, prices come from recent Aviasales searches and can change, booking links are affiliate links, an 8.5-hour ground cutoff drops and names airports, and a non-"ok" status must be relayed rather than prices invented.

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 routing are front-loaded and most sentences carry real instruction, but the block is very long and several rules are stated twice (dates semantics, the anywhere/warm-country handling, the max_ground_hours default already in the schema), and output-formatting guidance is interleaved with search guidance rather than grouped.

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 16-parameter tool with no output schema, the description still tells the agent what the answer contains (rows with origin/destination airports, notes, status, booking links, totals and currency) and how to present it. Nothing required to call it or interpret the result correctly is missing.

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 baseline is 3, but the description adds interpretation rules absent from the schema: a month or holiday without a year means the next one, don't ask for trip length, dates "assumed" vs "from_user" when the user gave no timing, and set mode only when the user states how they reach the airport. A worked example call anchors several parameters at once.

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 with scope: door-to-door round trips from a home town rather than from a single airport, ranking by price, total travel time, or balanced. It explicitly names the excluded case ("Not for a search from one airport already chosen"), so it is distinguishable from a plain flight search at a glance.

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 explicit triggers (user names a home town, several airports, or asks for cheapest/fastest), explicit exclusions (single airport already chosen), and per-case routing: one_way for one-way, leg-by-leg for multi-city with dates only, ask for dates otherwise. It also routes to siblings — trip_details when a trip is picked, price_calendar for cheapest-month questions, send_feedback for bad answers.

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 cover read-only/idempotent/non-destructive, and the description adds genuinely new context: prices come from observed fares, not live availability, and results are 'Instant'. It also discloses the ground-transport leg and per-person/home-currency pricing. Slightly short of 5 because it doesn't say how the estimates should be caveated for booking.

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-loaded with the core definition, then usage, then the hand-off, then prerequisites and units. Dense but every clause carries information; only the pricing-unit sentence trails slightly long.

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 carries the burden of explaining what comes back ('what fares on each route usually cost per month ... plus the train or bus to the airport'), which it does. It leaves the exact shape of the per-airport, per-month result implicit.

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 already 100%, so baseline is 3, but the description adds constraints the schema doesn't: month strings must be '2027-03' with at most 12 months, and prices default to the home currency unless `currency` is given. It also reinforces that 'home' is a town, not an airport.

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 ('Shows the cheapest month to travel from someone's home town to a place, for each airport within reach') and immediately narrows scope to fare history plus ground transport. An agent can distinguish it from find_trips without opening either 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?

Explicit trigger ('when the user's dates are flexible or they ask when it's cheapest to go') and an explicit hand-off ('then call find_trips for the month they choose to get bookable trips'). It also states a prerequisite: needs the home town, not an airport.

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 only say this is a non-read-only, non-destructive, closed-world, non-idempotent call. The description adds what annotations cannot: the report is transmitted to the product maker, is scoped to Doorfarer and never the user, and must exclude names, contact details, travel plans and personal disclosures. It also tells the agent to inform the user the report was sent, closing the loop on a write whose result is otherwise invisible with no output schema.

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 trigger conditions are front-loaded, followed by writing guidance, privacy rules and bookkeeping. Every sentence contributes, though the privacy paragraph and the 'be specific and honest' sentence overlap slightly and could be tightened into one clause.

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 supplies the crucial missing return behavior — tell the user it's sent — and fully specifies the privacy boundary and deduplication rule. It stops short of saying whether the report is attributed to the assistant or whether the maker may reply, a minor gap for a fire-and-forget feedback tool.

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 parameter definitions already carry the burden. The description still adds operational value by telling the agent where search_id comes from ('the answer's last line') and by constraining one report per search, plus guidance on what belongs in what_happened that the schema only paraphrases.

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?

Names a concrete verb-and-resource pair — send a report to Doorfarer's maker about where the product fell short — and frames its purpose as product feedback, which cleanly separates it from the read-only siblings find_trips, price_calendar and trip_details. An agent can tell instantly that this is the only outgoing-report tool in the set.

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 explicit trigger conditions: a find_trips answer that was wrong, confusing or missing information, or an explicit user request to pass something on. It also states the cadence constraint 'One report per search', which prevents duplicate filings, and defines the agent's voice when writing the report.

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, and non-destructive behavior, but the description adds meaningful context beyond them: local times, booking link, .ics calendar file, and the 72-hour retention window for trip ids. Those details materially affect whether and how an agent should call the tool.

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 description is front-loaded with the core behavior and then adds the necessary input source and retention constraint. Every sentence earns its place by clarifying what is returned or how to obtain the required id.

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 one required parameter, full schema coverage, and annotations covering safety traits, the description supplies the remaining operational context: what the output contains, how to obtain the id, and how long ids remain valid. No output schema is present, but the description sufficiently explains the return content.

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 baseline is 3. The description goes beyond the schema by explaining where the trip id comes from in the find_trips answer and by adding the important 72-hour retention constraint, though it does not add format or validation detail beyond that.

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 states a specific verb and resource: it shows one trip from a find_trips answer step by step. It clearly distinguishes this from the sibling find_trips by framing this as the detailed view of a selected trip, and it enumerates the journey segments the tool returns.

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 says when to use the tool: when the user picks a trip or asks how the journey works. It also routes the agent to the sibling find_trips as the source of the required trip id, which gives clear alternative-source guidance.

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