Les bancs

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, open-world), so the bar is lower. The description adds real behavioral value with no output schema present: it discloses the returned content (level-by-level, status, latest result) and the fact that carte scolaire publication status is reported, plus the arrondissement special case for Paris/Lyon/Marseille.

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?

Two sentences, front-loaded with the resource being returned, and the arrondissement caveat is efficiently appended. No filler, though the first sentence is dense and packs several concepts into one clause chain.

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 single-parameter lookup with no output schema, the description carries the needed burden: it says what is returned and the one tricky input case (arrondissements). Gaps remain around result granularity ('dernier résultat' of what?) and response shape, but the tool's complexity is low.

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% – code_insee is fully documented with a pattern and an example (49007 Angers) – so baseline would be 3. The description adds a parameter-relevant nuance by flagging that Paris, Lyon and Marseille count arrondissements, hinting that their codes may be arrondissement INSEE codes rather than the commune itself.

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 resource and scope: the establishments of one commune, broken down 'niveau par niveau' with their status and latest result, plus whether the collège/lycée catchment map is published. This clearly distinguishes it from the single-school siblings (get_french_school, get_french_school_ranking, get_french_school_sectors), though it never names those alternatives explicitly.

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 is only implied – an agent infers 'use this to list all schools in a commune'. The note that Paris, Lyon and Marseille count arrondissements is a genuinely useful usage nuance, but no explicit when-to-use/when-not or alternative tool is offered.

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, destructiveHint=false and openWorldHint, so the safety and repeat-call profile is covered structurally. The description's added value is disclosing the breadth of the returned payload (exam results across four sessions, IPS, rankings, Parcoursup formations), which helps an agent anticipate cost/response size. It does not address data freshness, coverage gaps, or error behavior for an invalid UAI.

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?

A single sentence with the identifying mechanism front-loaded ('Un établissement par son UAI'), followed by a dense field inventory. Every listed item is informative about the payload, though the run-on list is longer than strictly needed and could have been trimmed.

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?

There is no output schema, so the description carries the full burden of describing return content — and it does so comprehensively, covering identity, demographics, performance indicators, specialties, Parcoursup offerings and contact. What remains thin is operational detail: no statement on what happens for an unknown UAI, missing-data handling, or response shape.

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 100% and the schema already documents the UAI format with a pattern and example (0942434K). The description adds nothing beyond 'par son UAI', so it neither clarifies nor compensates for anything. Baseline 3 applies when the schema does the heavy lifting.

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+resource: it returns a single school record identified by its UAI, and it enumerates the fields returned (nature, statut, effectifs, IPS, exam results, valeur ajoutée, formations Parcoursup, contact). An agent can tell this is a single-entity lookup rather than a list. However, it never names or contrasts the siblings (get_french_commune_schools, get_french_school_ranking, get_french_school_sectors), and its mention of 'communes de son secteur' overlaps conceptually with the sectors tool, leaving some ambiguity.

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?

There is no when-to-use guidance, no prerequisites, and no alternatives named. The only implicit cue is 'par son UAI', which suggests the tool requires a known identifier, but the description never says to use this when you already have a UAI versus using the commune-based sibling when you do not. Usage must be entirely inferred from the sibling 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?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint, so the safety profile is covered. The description adds a genuine behavioral detail — the minimum candidate threshold that governs whether a school is ranked — but does not describe sort order behavior, pagination, or how ties/limits interact.

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?

Two sentences, front-loaded with the resource and scope, dense but free of filler. Every clause carries information (types, geography, output fields, eligibility threshold), though the length of the first sentence slightly taxes readability.

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 usefully enumerates the returned fields (rank, name, commune, score, success rate, mentions). For a 5-parameter tool with only 40% schema coverage, however, it leaves limit and sector unexplained and offers no interaction details, so it is adequate but not complete.

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 40%: code_insee and department carry descriptions, while type, limit and sector do not. The description partly compensates by glossing each type value (collèges/brevet, lycées/bac, lycées pro/bac pro) and by naming the three geographic scopes, which hints at the code_insee vs department distinction. It adds nothing for limit or sector.

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 resource (school rankings) and enumerates the three school types it covers, mapping each to the exam it corresponds to (brevet, bac, bac pro). It also specifies the geographic levels. However, it never references the sibling tools (get_french_school, get_french_commune_schools, get_french_school_sectors), so the agent must infer differentiation on its own.

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 is implied through the scoping options (France, department, commune) and the eligibility threshold (40 candidates for France, 20 elsewhere), which helps the agent judge what will return results. But there is no explicit when-to-use vs when-not, and no routing to the sibling tools that return individual school or sector data.

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, openWorldHint, idempotentHint and destructiveHint=false, so the safety profile is covered externally. The description adds value on result content (carte scolaire publiée, seule école publique, secteur non publié) but says nothing about error handling, address-not-found behavior, or latency/limits, which are the traits annotations do not cover.

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 core purpose (sector establishments, level by level) before the supporting detail, and the sibling-routing note closes it cleanly. Dense, but each clause carries content rather than 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 carries the burden of describing returns and does so thoroughly (per-level sector, nearby public, private under contract, post-bac/Parcoursup). Only failure/edge-case behavior is left unstated, a minor gap for a single-parameter read 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% with a single, well-documented address parameter that even includes an example. The description adds no syntax, format or constraint detail beyond the schema, so the baseline 3 applies.

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 resource (sector establishments for a French address), scoped per level (maternelle/élémentaire, collège, lycée), and enumerates what the result contains. It explicitly differentiates from the sibling get_french_commune_schools by noting a commune has no sector, so an agent can pick between them without opening a 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 a clear routing rule: for a commune rather than an address, use get_french_commune_schools. That is a real when-to-use-this-vs-alternative statement. It stops short of stating exclusions or prerequisites (e.g. address must be a valid French address, what happens on failure).

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