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

With no annotations, the description carries full burden and excels: it discloses the geographic coverage limitation, the model-forecast nature, the possibility of an empty response due to stale/incomplete runs, and that an out-of-coverage notice differs from a blank response. This goes well beyond basic behavior.

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 dense but well-structured: a one-line summary, followed by output specifics, coverage/limitations, edge-case behavior, and parameter notes. Every sentence adds relevant information without redundancy, achieving conciseness through organization rather than brevity for its own sake.

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?

Given the tool's moderate complexity and the existence of an output schema (which covers return values), the description covers all necessary context: what it returns, geographic coverage, data source caveats, how to handle edge cases, and parameter preparation. It leaves no obvious gaps for an agent to misuse the 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 0%, so the description must compensate. The Args section provides decimal examples for latitude/longitude (Vienna coordinates) and explicitly states 'Geocode city names to coordinates yourself,' giving meaningful semantic guidance beyond the bare number type in the schema.

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 opens with a specific verb and resource: 'Get air quality for a location: pollutants now and the AQI outlook.' It further details the exact pollutants (NO₂, O₃, PM10, PM2.5) and the European Air Quality Index bands, making it unmistakably distinct from sibling weather/storm tools.

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 provides clear context and restrictions: 'serves Austria and the Alpine region only,' 'model forecasts, not station measurements,' and instructs users to geocode city names themselves. However, it does not explicitly name alternatives or state when not to use this tool versus its siblings, stopping short of a 5.

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?

With no annotations, the description carries the full burden and does well: it names the underlying datasets (AROME/INCA/nowcast), discloses that out-of-grid requests yield an out-of-coverage notice rather than an error, and says the response states which datasets served it. It omits latency, caching, or rate-limit behavior, which keeps it below a 5.

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 the verb and resource, then layers scope, provenance, and semantics in short paragraphs; each block is load-bearing. The brief note on how falling precipitation is decided is slightly tangential but reinforces trust in the result, so it earns most of its space.

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 an output schema already present, the description covers everything an agent needs: what it returns conceptually, geographic limits, failure mode outside coverage, and parameter format. Nothing material is left to inference.

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 0% and the schema supplies only bare 'number' types, so the description does the real work: it specifies decimal latitude/longitude semantics, gives concrete Vienna values for both fields, and tells the agent to resolve place names itself. Without this text the parameters would be effectively undocumented.

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 current weather conditions') and immediately scopes it by data source and geography. The word 'current' plus the observation/symbol explanation clearly separates it from siblings like get_hourly_forecast and get_air_quality without needing to name them.

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 a hard applicability rule (Austria and the Alpine region only, out-of-coverage point returns a notice) and an explicit responsibility for the caller (geocode city names to coordinates yourself). It stops short of routing advice such as when to prefer get_hourly_forecast or get_storm_outlook instead.

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?

With no annotations at all, the description carries the full behavioral burden, and it does disclose valuable non-obvious traits: the precipitation chance is a C-LAEF ensemble-derived estimate (0/30/70/95 %) rather than a published probability, coverage is limited and out-of-grid points yield a notice, and times are Vienna local. It does not describe errors, ordering, or rate behavior, so it is solid but not exhaustive for an annotation-free 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?

Front-loaded with the one-line purpose, then a compact paragraph of caveats, then per-argument notes in a conventional Args block. The provenance paragraph is long but each sentence (resolution, symbol source, precipitation derivation, coverage, timezone) carries information an agent cannot get from the schema.

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 4-parameter read tool with an output schema (so return fields needn't be explained), the description covers the remaining unknowns: data source and resolution, forecast horizon, geographic coverage and failure mode, timezone convention, and clamping. Nothing material is left for the agent to guess.

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 0%, and the description fully compensates: it explains latitude/longitude with concrete Vienna examples, hours with its default and 1–60 clamp, and start as an optional ISO 8601 instant with the crucial detail that a missing offset means Vienna local time matching output times.

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 ('hour-by-hour weather forecast for a location') and immediately narrows the product: GeoSphere AROME at 1 km resolution, up to ~60 h. Combined with sibling names like get_current_weather, an agent can tell this is the multi-hour forecast tool rather than a current-conditions or air-quality tool.

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 the operative preconditions for calling it: coordinates must be supplied by the caller ('Geocode city names to coordinates yourself') and the point must fall in the Austria/Alpine grid or an out-of-coverage notice is returned. It does not explicitly say 'use get_current_weather instead when you only need present conditions', so routing to siblings is left implied by names.

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?

With no annotations, the description carries the full burden and does substantially well: it discloses the coverage boundary and the out-of-coverage notice, the definition logic for a thunderstorm hour, stale-run handling, truncated horizons, and the "already in progress" timestamp semantics. It omits units for peak gust and CAPE (km/h vs m/s, J/kg is stated only for CAPE thresholds) and says nothing about error or failure behavior.

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 a one-line summary, then structured into return content, detection logic, data sources/coverage, horizon semantics, and args. It is on the long side, and the opening enumeration of return values partly duplicates what the output schema already declares — that portion is the weakest use of space.

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 tool with an output schema, the description covers what the schema cannot: detection criteria, coverage limits, horizon rounding, staleness, and the deliberately verdict-free contract. Missing units and any error-mode detail are the remaining gaps, but nothing needed to call it correctly 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 0%, so the description is the only source of parameter meaning — and it supplies it: decimal degrees with concrete Vienna examples, plus the explicit instruction to geocode city names to coordinates yourself, and a note that points outside the grid are rejected. It does not state valid ranges or what happens with malformed input, but the essential semantics are covered.

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?

Opens with a specific verb and resource ("Get the severe-weather outlook") and immediately enumerates the distinct outputs — peak gust, thunderstorm presence/timing, peak CAPE — which separates it from get_current_weather and get_hourly_forecast in substance. It never names the siblings explicitly, so an agent must infer that this is the gust/storm-specific tool rather than a general forecast.

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?

Usage context is implied rather than stated: it makes clear the tool is diagnostic-only ("Deliberately reports no severity verdict") and geographically restricted to Austria/Alpine, which does route some callers away. But there is no explicit comparison to get_hourly_forecast or get_current_weather — the agent is left to infer that hourly conditions belong to the sibling and storm/convective signals belong here.

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