Logements neufs des promoteurs partenaires de M2 : programmes, lots, prix et disponibilités.
Scored across 7 tools
Each tool has a fairly distinct purpose: search returns a list, get_lot returns a full record, get_program covers a residence, compare_lots builds a side-by-side table, and create_lead submits contact details. Some overlap exists around availability, because status and freshness also appear in search_lots, get_lot and compare_lots, so check_availability is not the only way to confirm whether a lot is available. The descriptions clarify that check_availability refers to a dedicated verification action, which keeps the tools distinguishable.
All seven names follow the same snake_case verb_noun pattern: check_availability, compare_lots, create_lead, get_lot, get_program, list_promoters, search_lots. The verbs vary according to the action, but that is expected and still predictable, and no naming style is mixed. There are no vague or duplicated names.
Seven tools are well within the ideal range for a real-estate catalog assistant. Each one supports a distinct step of the buyer journey: finding lots, viewing details, reading program information, filtering by promoter, comparing, verification, and making contact. No tool seems redundant.
The tool surface covers the full buyer journey, from property discovery through lead handoff, and there are no dead ends. Minor gaps remain: there is no standalone list_programs tool, so program IDs must be obtained through search_lots or a lot record, and there is no way to check the status of a previously created lead. Agents can work around these limitations.
Vérifie le statut commercial actuel (available disponible, on_hold optionné, reserved réservé, sold vendu) d'un ou plusieurs lots, avec la date de dernière mise à jour transmise par le promoteur et un indice de fiabilité (high < 24 h, medium < 72 h, low au-delà).
Utiliser avant de recommander un lot ou de proposer une mise en relation, et quand l'acquéreur demande si un logement est toujours disponible.
| Name | Required | Description | Default |
|---|---|---|---|
| lot_ids | Yes | Identifiants des lots à vérifier. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
readOnlyHint=true already tells the agent this is a safe read, yet the description still adds genuine behavioral context the annotations cannot: results carry the promoter's last-update timestamp and a freshness grade with concrete thresholds (high < 24 h, medium < 72 h, low beyond). It does not discuss pagination or what happens if a lot id is unknown, 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?
Two sentences, front-loaded with the action and its result shape, and every clause carries information: status enum glosses, the update date, the reliability scale with thresholds, then the usage triggers. Nothing is redundant or restated 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?
With no output schema, the description compensates by documenting in prose what comes back (status values, update date, reliability index) and when to call it. An agent has enough to invoke it correctly and to interpret the response without further documentation.
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 single parameter is already documented in the schema, so the baseline of 3 applies. The phrase 'un ou plusieurs lots' confirms batch behavior, but the description never mentions the schema's maxItems=20 cap or what partial failures look like.
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 ('Vérifie le statut commercial actuel ... d'un ou plusieurs lots') and enumerates the four possible status values with glosses, so an agent knows exactly what the call answers. It also distinguishes the payload from a plain lot fetch by naming the last-update date and a reliability index as part of the result.
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 two concrete triggers: before recommending a lot or proposing a connection, and when the buyer asks whether a property is still available. There is no explicit exclusion or sibling routing (e.g. 'use get_lot for full details, search_lots for discovery'), so the guidance is contextual but not exhaustive.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Affiche un tableau comparatif de 2 à 4 lots (prix, surface, extérieur, étage, livraison, points forts, statut et fraîcheur).
Pour formuler une recommandation, renseigner recommended_lot_id avec le lot le plus adapté au projet exprimé par l'acquéreur (résidence principale, investissement, budget…) et reasons avec 2 à 4 justifications courtes, fondées uniquement sur les données des lots comparés. La recommandation est mise en évidence dans le tableau.
Le tableau se termine par un formulaire de mise en relation (logement, nom, e-mail, téléphone) qui transmet directement la demande au conseiller : ne pas redemander ces informations dans la conversation, inviter l'acquéreur à utiliser le formulaire.
Le tableau est affiché directement à l'utilisateur : ne pas le recopier dans la réponse.
| Name | Required | Description | Default |
|---|---|---|---|
| lot_ids | Yes | Identifiants des lots à comparer, dans l'ordre d'affichage. | |
| reasons | No | Raisons courtes justifiant la recommandation, ex. « Un bon équilibre entre prix, surface et terrasse ». | |
| recommended_lot_id | No | Identifiant du lot recommandé pour le projet de l'acquéreur (optionnel). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations only declare readOnlyHint=true, and the description adds substantial non-obvious behavior: the table embeds a contact form that transmits a lead directly to the advisor, the table is rendered to the user and must not be copied into the response, and the recommendation is highlighted in the table. These disclosure points materially change how the agent should respond.
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?
Four short paragraphs, each front-loading a distinct concern (output content, recommendation inputs, contact form, rendering rule), with no filler. It is slightly dense but every sentence carries operational weight.
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, yet the description fully describes what the user will see (a comparison table with named dimensions, highlighted recommendation, and a lead-capture form) and tells the agent how to behave around that output. Nothing needed to invoke or narrate the result 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 baseline is 3, but the description adds real meaning: lot_ids order becomes display order, recommended_lot_id must be tied to the buyer's stated project (résidence principale, investissement, budget), and reasons must be grounded strictly in the compared lot data. The 2-4 count for reasons is only in the schema, so this is a modest addition rather than a full elaboration.
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 plus scope: 'Affiche un tableau comparatif de 2 à 4 lots', and enumerates the compared dimensions (prix, surface, extérieur, étage, livraison, points forts, statut, fraîcheur). This is clearly distinct from single-lot retrieval (get_lot) or listing (search_lots) among the siblings, even without naming 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 explicit downstream instructions: fill recommended_lot_id with the lot best matching the buyer's project, provide 2-4 short reasons grounded only in compared lot data, and never re-ask contact details in chat. It never states when to prefer this over get_lot or search_lots, so the alternative-selection guidance is implicit rather than explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Transmet une demande de mise en relation avec un conseiller commercial du promoteur, à propos d'un lot ou d'un programme. N'appeler qu'après accord explicite de l'acquéreur, avec les coordonnées qu'il a lui-même fournies dans la conversation (ne jamais les inventer). Le conseiller recontacte l'acquéreur ; aucun rendez-vous n'est confirmé par cet outil.
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | Nom et prénom de l'acquéreur. | |
| Yes | E-mail de l'acquéreur. | ||
| phone | No | Téléphone de l'acquéreur. | |
| budget | No | Budget d'achat TTC en euros, s'il a été exprimé par l'acquéreur. | |
| lot_id | No | Lot concerné. Requis si program_id est absent. | |
| channel | No | Origine de la demande : `conversation` (défaut) ou `comparison_form` (formulaire du comparatif). | |
| message | No | Contexte utile au conseiller (critères, questions, financement…). | |
| purpose | No | Nature du projet d'achat. | |
| incentive | No | Dispositif envisagé par l'acquéreur (PTZ, TVA réduite, LMNP), s'il l'a mentionné. | |
| program_id | No | Programme concerné. Requis si lot_id est absent. | |
| preferred_callback_time | No | Disponibilités de l'acquéreur pour être rappelé, ex. « mardi après-midi ». |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations carry only a title, so the description must do the work, and it does: it discloses that this is a submission to a human advisor, that the advisor recontacts the buyer, and that no appointment is confirmed by this tool. It does not address duplicate handling, idempotency, or error behavior for an 11-parameter write.
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, no filler, with the action stated first and the consent precondition immediately after. Every clause carries an operational constraint.
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 mutation tool with no output schema and no behavioral annotations, it covers purpose, precondition, data-source rule and downstream effect. It is close to complete; only duplicate/idempotency and what the tool returns on success are left unstated.
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 each of the 11 parameters, including the three enums, is documented in the schema. The description only adds the lot-or-program framing and the sourcing rule for the contact fields, which the schema already conveys via required-if-absent wording, 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 verb and resource: transmits a contact request to a promoter's sales advisor regarding a lot or a program. This is unmistakably distinct from the read-only siblings (get_lot, search_lots, compare_lots), though the description never names or contrasts them 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?
Gives an explicit precondition (only after the buyer's explicit consent, with contact details the buyer supplied in the conversation) and an explicit prohibition (never invent the contact details). It also states what happens next, so an agent knows the call does not book anything.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Renvoie la fiche complète d'un lot : surface, étage, extérieur, orientation, parking, plan, prix, statut et fraîcheur de la donnée, ainsi que le programme (adresse, livraison, prestations, points forts, dispositifs éligibles). Utiliser quand l'acquéreur veut « voir le détail » d'un logement.
| Name | Required | Description | Default |
|---|---|---|---|
| lot_id | Yes | Identifiant du lot (champ `id` renvoyé par search_lots). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
readOnlyHint=true already establishes the safety profile, so the description's job is to add context — and it does, notably by disclosing that the payload includes data freshness/recency and a nested program block. It does not cover failure behavior for an invalid lot_id or any access constraints, which keeps it from 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?
Two sentences, zero filler: the first front-loads the full return contract as a compact enumeration, the second gives the usage trigger. Nothing is redundant.
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 return contents and does so thoroughly, including nested program data. Only the error case for a non-existent lot_id is left unspecified.
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 the schema itself explains that lot_id is the `id` returned by search_lots, so semantics are already fully documented. The description adds no additional meaning about the parameter, which is the expected baseline when the schema does the work.
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 ('Renvoie la fiche complète d'un lot') and enumerates the exact contents returned, so an agent knows precisely what comes back. It implicitly separates itself from search_lots (list) and compare_lots (comparison), but the overlap with get_program — since program details are included here — is not addressed.
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 concrete trigger condition: use when the buyer wants to 'see the detail' of a dwelling. It provides clear contextual guidance but no exclusions or explicit alternatives (e.g. when to prefer get_program or compare_lots instead).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Renvoie la présentation d'un programme immobilier neuf (promoteur, localisation, description, livraison, prestations, points forts, dispositifs éligibles) et la liste de ses lots disponibles. Utiliser quand l'acquéreur s'intéresse à une résidence dans son ensemble plutôt qu'à un lot précis.
| Name | Required | Description | Default |
|---|---|---|---|
| program_id | Yes | Identifiant du programme (champ `program.id` d'un lot). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, so the safety profile is covered. Beyond that, the description discloses the substantive return content: programme presentation fields and the list of available lots, which is valuable given there is no output schema. It does not cover error behaviour such as an unknown program_id, but the core behaviour is well communicated.
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 with no waste: the first front-loads what is returned, the second front-loads when to use it. Every part earns its place.
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, read-only lookup with no output schema, the description supplies both the return content and the selection context. It is nearly complete, with only minor gaps around failure modes or authentication requirements.
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 single parameter is already documented in the schema as the programme identifier tied to a lot's program.id. The description adds no further parameter semantics, so the baseline of 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 uses a specific verb ('Renvoie') with a clearly scoped resource: the full presentation of a new-build property programme plus its available lots. It enumerates the returned fields (promoteur, localisation, description, livraison, prestations, points forts, dispositifs éligibles), making the tool distinguishable from sibling lot-level tools like get_lot.
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 an explicit usage condition: use when the buyer is interested in the residence as a whole rather than a specific lot. This effectively routes between programme-level and lot-level queries, though it does not name the alternative sibling tool (e.g. get_lot) directly.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Liste les promoteurs partenaires de M2 présents dans le catalogue, avec leur nombre de programmes et de lots disponibles. Utiliser pour identifier un promoteur cité par l'acquéreur avant de filtrer search_lots.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
readOnlyHint=true already declares the safety profile, so the description's job is to add context — which it does by disclosing the payload shape (each promoter with its program and available-lot counts) and the implicit scope limit (only M2 partner promoters in the catalog). No auth, rate-limit, or ordering/pagination behavior is mentioned.
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, no filler, with the resource and return content front-loaded and the routing hint second. Every clause earns its place.
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 zero-param read tool with no output schema, the description covers what is listed, what is returned, and when to call it. Minor gaps remain: whether the full catalog is returned at once and in what order, which matters since there is no output schema to fall back on.
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?
The tool takes zero parameters, so there is nothing for the description to disambiguate; baseline 4 applies. It correctly avoids inventing filtering options the schema does not expose.
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 ("Liste les promoteurs partenaires de M2 présents dans le catalogue") and adds what each entry carries (program count, available lot count). It also names the sibling it feeds into (search_lots), so the agent can place it in the workflow without opening another 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?
The second sentence gives a concrete trigger: use it to identify a promoter the buyer mentions, before filtering search_lots. That is a clear when-to-use plus an alternative-tool relationship. It stops short of stating exclusions (e.g. when to go straight to search_lots or get_program instead).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recherche des logements neufs (lots) dans le catalogue des programmes partenaires de M2, et les affiche sous forme de liste.
Utiliser dès qu'un acquéreur décrit un projet : ville, typologie (T2, T3...), budget, surface, extérieur (terrasse, balcon...), date de livraison, promoteur.
Par défaut seuls les lots disponibles sont renvoyés, triés par prix croissant. Chaque lot porte un indicateur de fraîcheur : si reliability vaut low, prévenir l'utilisateur que la disponibilité doit être confirmée (check_availability).
Ne jamais inventer de logement absent des résultats.
Ne jamais retirer ni assouplir un critère de l'acquéreur de sa propre initiative : en l'absence de résultat, présenter les pistes d'élargissement renvoyées (relaxations) et laisser l'acquéreur choisir.
La liste est affichée directement à l'utilisateur, sans bouton d'action : ne pas la recopier dans la réponse, mais proposer explicitement la suite : le détail d'un logement (get_lot), la comparaison de 2 à 4 logements (compare_lots, en demandant lesquels comparer) ou une mise en relation.
| Name | Required | Description | Default |
|---|---|---|---|
| city | No | Ville recherchée, ex. « Lyon ». | |
| limit | No | Nombre maximum de lots renvoyés (1 à 20, 6 par défaut). | |
| district | No | Arrondissement (Paris, Lyon, Marseille), ex. 7 pour Lyon 7e. | |
| min_area | No | Surface habitable minimum en m². | |
| promoter | No | Nom ou identifiant (`slug`) du promoteur, ex. « LNC ». Voir list_promoters. | |
| incentive | No | Dispositif d'aide ou fiscal éligible. | |
| max_price | No | Prix TTC maximum en euros (budget). | |
| min_price | No | Prix TTC minimum en euros. | |
| unit_types | No | Typologies acceptées. | |
| outdoor_spaces | No | Types d'extérieur acceptés. | |
| min_outdoor_area | No | Surface minimum de l'extérieur en m². | |
| include_unavailable | No | Inclure les lots optionnés, réservés ou vendus. Faux par défaut. | |
| latest_delivery_date | No | Date de livraison maximale (AAAA-MM-JJ). « Livrable en 2027 » → 2027-12-31. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The annotations only declare readOnlyHint, but the description adds substantial behaviour: available-only filtering by default, ascending price sorting, a reliability indicator that can trigger check_availability, a prohibition on inventing or relaxing criteria, and UX rules about displaying the list directly without copying it. These are concrete operational constraints beyond the annotations.
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, then usage trigger, defaults, reliability handling, safety rules, and next-step behaviour. The length is justified by the 13-parameter scope and high-stakes UX constraints; each sentence carries a distinct rule.
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 read-only search tool with no output schema, the description covers the trigger, default filtering, sorting, reliability handling, no-result relaxations, anti-hallucination rules, and follow-up actions. Nothing essential for correct invocation or user-facing behaviour 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%, and all 13 parameters already have descriptions, defaults, formats, or enums. The description groups buyer criteria into search concepts and mentions default availability and sorting, but adds little per-parameter syntax or meaning beyond the schema, so the baseline 3 is appropriate.
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?
Starts with a specific verb, 'Recherche', and a specific resource, 'logements neufs (lots)', in the partner catalogue, and states that it returns a list. It also distinguishes this broad search from sibling actions such as check_availability, get_lot, compare_lots, and list_promoters.
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: 'Utiliser dès qu'un acquéreur décrit un projet', followed by the buyer criteria city, typology, budget, surface, outdoor space, delivery date, and promoter. It also states the default available-only behaviour, forbids relaxing criteria unilaterally, and routes follow-up actions to get_lot, compare_lots, contact, or check_availability when reliability is low.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
check_availabilitycompare_lotscreate_leadget_lotget_programlist_promoterssearch_lotsPrix au m² communal, synthèse de marché et formalité d'urbanisme, sur données publiques françaises.
French real-estate intelligence: prices, rents, listings, yield, rental management.
Estimation DVF indicative, frais de notaire et rendement locatif, en lecture seule.
Alertes Hunter, analyse d'un bien (écart au prix payé, loyer, verdict), quartiers classés par IMS.
Add one secure layer between your agents and this server.
