France Care, accès aux soins et prévention

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

Annotations already declare readOnlyHint, idempotentHint and non-destructive, so the safety profile is covered. The description adds genuinely new behavioral context: the input disambiguation priority (unprefixed five digits read as postal code first) and the ambiguity-resolution flow that returns multiple communes with a relaunch query. It stops short of describing pagination or latency, but no output schema exists and the return content is summarized.

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?

Content is front-loaded: purpose, accepted input forms, ambiguity handling, then return contents and sources. It is dense and long, but since no output schema exists, the enumerated return fields and source/cadence notes largely earn their place rather than repeating structured metadata.

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 complex domain tool with one parameter and no output schema, the description supplies everything an agent needs: accepted identifier formats, ambiguity fallback, a detailed account of returned data, and the upstream sources with millesime and refresh cadence. Nothing material 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% and there is only one parameter, so the baseline would be 3. The description goes beyond the schema by specifying the disambiguation order ('sans préfixe, 5 chiffres se lisent d'abord comme un code postal') and richer examples, actively clarifying how a single string argument is interpreted.

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 ('Analyse l'accès aux soins d'un territoire français') and enumerates exactly what is returned, so an agent knows it is a single-territory healthcare-access analysis. It does not explicitly distinguish itself from the sibling compare_territories, which is left to inference from 'un territoire'.

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 thoroughly covers how to supply input (postal code, commune name, INSEE prefix) and how to recover from ambiguity via relancer_avec, which implies usage. However, it never states when to prefer this over siblings such as compare_territories, search_health_infrastructure or qui_saisir, and gives no exclusions.

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 closed-world, so safety is covered. The description adds genuine behavioral context beyond that: only validated sheets are returned, no personal data is involved, and results carry a verification date and source link, which speaks to data provenance and trustworthiness.

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 what the tool returns, followed by when to use it, then short constraint fragments. Dense and information-rich, though the long enumeration of sheet fields in the first sentence borders on over-listing rather than crisp conciseness.

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?

All three optional parameters are documented at 100%, the annotations carry the safety profile, and although there is no output schema, the description itself enumerates the returned fields and notes the validation/no-personal-data constraints. Nothing essential for correct invocation 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 description coverage is 100%, so the baseline is 3. The description goes beyond the schema by explaining the empty-argument behavior ('Sans argument : la liste des thématiques') and by mapping theme and month queries to the sujet/mois parameters with concrete examples, adding real meaning to how the parameters are used.

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 (the France Care guide sheets of workplace prevention animations) and enumerates exactly what each sheet contains (problem solved, step-by-step flow, cost, preparation, partners, vigilance, published efficacy, sources, verification date, link). This is clearly distinguishable from sibling tools such as analyze_territory, calendrier_sante or search_health_infrastructure, which concern territory or infrastructure data.

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 states the trigger context ('quand un employeur, un DRH, un CSE ou un préventeur cherche une idée d'action de prévention') and the two query modes (by theme or by month), plus the no-argument fallback. It gives clear when-to-use context but does not name or exclude any alternative sibling tool.

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 a safe, idempotent, closed-world read. The description adds useful data-level context the annotations cannot: dates may be 'à confirmer', coverage scope is France/Europe/world, and an organizer plus a cross-link to guide animations is included.

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 resource and publisher are front-loaded, followed by the field list and the usage sentence; no sentence is filler. The middle field enumeration is dense but justified since there is no output schema to carry that information.

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 compensates by describing the shape of returned entries, and annotations cover the safety profile while the schema covers all parameters. Nothing an agent needs to call it correctly appears to be 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 mois, annee and sujet are already fully documented with defaults (all-year, current/next year, keyword matching). The description only alludes to month filtering via the example, adding no syntax or format 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?

The description names a specific resource (health days/weeks/months calendar) with its publisher (France Care) and enumerates the returned fields, so the agent knows exactly what it retrieves. It does not explicitly contrast with any sibling tool, but the resource is distinctive enough that confusion is unlikely.

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 gives concrete usage contexts — preparing a yearly prevention plan, internal communication, or finding which event falls in a given month — which orients the agent well. It stops short of naming alternatives or exclusions, so it is clear context without routing rules.

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=false, so the safety profile is fully covered by structured data. The description adds only the accepted input formats and the exact-2 constraint, which is useful but not deep behavioral context.

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, front-loaded sentences with no filler; the verb+resource leads and format details follow. The final sentence partly restates the schema's min/maxItems constraint, a minor redundancy.

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 simple, read-only, single-parameter comparison tool with no output schema, the definition covers inputs and scope adequately. It stops short of hinting at what the comparison returns (metrics/fields), but annotations cover the safety dimension.

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 documented array parameter, so the schema already carries the semantics. The description reinforces the format and supplies a concrete example ('insee:77186') but adds little beyond the schema's own parameter description.

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 (Compare) and resource (acces aux soins entre 2 territoires francais), making the pairwise scope explicit. It is implicitly distinguishable from the singular analyze_territory sibling via 'entre 2 territoires', but never names the alternative directly.

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 'exactly 2 territories' constraint implies when to use this rather than the single-territory analyze_territory, but the description never states the when/when-not or names the alternative. Usage is implied rather than guided.

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 read-only, idempotent, closed-world and non-destructive, yet the description adds substantial extra context: what the response contains (interlocutors with roles, dated Légifrance links, official fiches, verbatim deadlines, FINESS coordinates), how to consume it (refer to a_lire_d_abord, cite each deadline with article and verification block), and how to invoke it (choose a besoin, call argument-free for the list). The privacy constraint is an important behavioral disclosure beyond structured fields.

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 is front-loaded and the actionable instructions (privacy rule, use the fiches, cite deadlines) all earn their place. The long parenthetical enumeration of interlocutors makes the middle dense and run-on, but it conveys genuine scope 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 fully covers what is returned and how to use it, plus invocation rules and privacy constraints. Nothing an agent needs to call and correctly apply this tool 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 both parameters are documented structurally. The description adds usage meaning beyond the schema: 'Choisir le besoin dans la liste (appeler sans argument pour la description de chacun)' explains the no-argument behavior and the enum's purpose, and the lieu handling is reinforced as optional context for local coordinates and ARS.

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 precise verb+resource statement: who to contact, in what order, and on what legal text, for health/disability/loss-of-autonomy procedures in France. It enumerates the specific interlocutors (DAC, MDPH, ARS, juge, etc.) and the assets returned (roles, dated Légifrance texts, service-public.fr fiches, deadlines, FINESS coordinates), which cleanly distinguishes it from siblings like signaler_besoin_non_couvert.

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 gives explicit trigger conditions ('quand une personne, un aidant ou un professionnel ne sait pas à qui s'adresser, veut contester une décision ou faire une réclamation, ou prépare un courrier'), which is clear when-to-use guidance. It also states a hard privacy exclusion (never transmit the person's situation, name or health status), though it does not name sibling alternatives to route away to.

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, so the safety profile is covered. The description adds real behavioral context beyond that: it discloses the structure of retour for EHPAD results (offre_hebergement with places, habilitation, prix CNSA, HAS quality class). It does not discuss pagination or limits behavior, keeping it at 4.

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 purpose before modes, then uses parentheses and a numbered list for the two search strategies. The closing EHPAD paragraph on retour fields is dense but informative rather than redundant, so little is wasted.

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 6 parameters, 100% schema coverage, no output schema, and read-only annotations, the description supplies everything needed: the resource, the two modes, the relevant filters, and a substantive sketch of what results contain for the key use case (EHPAD).

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 further by explaining the relationship between radius_km and search mode and by motivating the alzheimer and aide_sociale filters, adding selection logic that the raw schema entries state only individually.

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 ('Recherche') and resource ('etablissements de sante dans la base FINESS'), enumerates the covered establishment types, and lays out two distinct search modes. An agent instantly knows this is a FINESS facility lookup and can tell it apart from territory-analysis siblings like analyze_territory and compare_territories.

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 when-to-use guidance for the proximity mode: 'Utilise radius_km quand l'utilisateur demande les etablissements les plus proches... dans un rayon de X km', and clarifies the default administrative-limits mode. No explicit when-not or named alternative tool, but the mode-selection context is clear.

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?

The description adds meaningful behavioral context beyond the annotations: no personal data is stored, inputs are restricted to closed lists, no free text or identity is accepted, and France Care aggregates reports by territory. These are useful operational details that annotations alone do not convey.

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 cover what the tool does, when to use it, and key input constraints. It is front-loaded with the core purpose and contains no redundant or filler content.

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 simple record-type tool with three fully documented required parameters and no output schema, the description gives sufficient context: purpose, usage trigger, privacy behavior, and aggregation. It does not describe the response format, but that gap is minor given the tool's simplicity.

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%, so the schema already documents commune, profession, and manque fully. The description reinforces that values come from closed lists and that no free text is allowed, but it does not add substantial new meaning beyond what the schema already provides.

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 uses a specific verb ('Enregistre') and a clear resource ('qu'une personne n'a pas trouve de soignant pour un besoin donne dans sa commune'). It clearly distinguishes this reporting tool from the analysis/search siblings by focusing on recording a signalement rather than analyzing or searching.

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 states when to use it: 'A utiliser quand la personne demande a signaler ce manque ou l'accepte.' It provides clear context for invocation, though it does not explicitly list alternatives or exclusion conditions relative to sibling tools.

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