Skip to main content
Glama

TOISE — intelligence foncière

Propositions IA sur un terrain — sentiment, brouillon d’accord, résumé (lecture)

toise_sentiment_proprietaire
Read-only

Que propose l'IA sur un terrain de mon portefeuille : sentiment d'un propriétaire, brouillon d'accord, résumé — jamais enregistré. Modes sentiment, brouillon, resume, courrier (brouillon à valider dans le CRM) ; rien n'est enregistré ni envoyé.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "mode": "<mode>",
      "terrain_id": "<terrain_id>"
    },
    "name": "toise_sentiment_proprietaire"
  }
}

Input Schema

NameRequiredDescriptionDefault
modeYes
sirenNo
terrain_idYes

Output Schema

NameRequiredDescriptionDefault
viaNo
modeYes
droitNo
liensNoLiens vers la fiche TOISE et vers des services de…
motifNo
objetNo
quotaNoQuota du jour après cet appel : { fonction, limite…
sirenNo
envoyeNoToujours false.
lignesNo
raisonNo
curseurYes
montantNo
partielNotrue : une source n’a pas répondu, des champs manquent…
partiesNo
sourcesYes
citationNoPhrase recopiée mot pour mot d’une source…
courrierNoMode courrier : texte du brouillon, à relire et valider…
has_moreYes
intituleNo
warningsYes
a_validerNoToujours true : proposition de l’IA, non enregistrée, à…
brouillonNo
parcellesNo
sentimentNo
a_discuterNo
enregistreNoToujours false : aucun accord n’est créé par cet outil.
sentimentsNoMode brouillon refusé : sentiment retenu…
terrain_idNo
quota_unitesNoUnités d’usage du jour : consommees_par_cet_appel…
montant_motifNo
champs_absentsNoDonnées absentes du résultat partiel.
statut_proposeNo
lien_validationNo
limite_atteinteNoPrésent quand un quota, une limite d’offre ou une fonction…
paiement_requisNoPrésent quand le résultat complet est un achat à l’acte…
conditions_notesNo
montants_evoquesNo
prochaine_actionNo
champs_a_completerNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed7 schema fields changed
    • changedInput schema / properties / mode / enum
      Previous value: -[
      -  "sentiment",
      -  "brouillon",
      -  "resume"
      -]New value: +[
      +  "sentiment",
      +  "brouillon",
      +  "resume",
      +  "courrier"
      +]
    • addedOutput schema / properties / champs_a_completer
      Added value: +{
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedOutput schema / properties / courrier
      Added value: +{
      +  "description": "Mode courrier : texte du brouillon, à relire et valider…",
      +  "type": "string"
      +}
    • addedOutput schema / properties / envoye
      Added value: +{
      +  "description": "Toujours false.",
      +  "type": "boolean"
      +}
    • addedOutput schema / properties / lien_validation
      Added value: +{
      +  "type": "string"
      +}
    • changedOutput schema / properties / mode / enum
      Previous value: -[
      -  "sentiment",
      -  "brouillon",
      -  "resume"
      -]New value: +[
      +  "sentiment",
      +  "brouillon",
      +  "resume",
      +  "courrier"
      +]
    • addedOutput schema / properties / objet
      Added value: +{
      +  "type": "string"
      +}
  2. First observed

TDQS

B3.4/5.0
Behavior4/5

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

Annotations already declare readOnlyHint=true and destructiveHint=false, and the description reinforces that with 'jamais enregistré' and 'rien n'est enregistré ni envoyé', so the safety profile is unambiguous. It also adds workflow context the annotations don't carry: the draft is an unvalidated proposal that must be confirmed in the CRM. It does not explain why idempotentHint=false for a read-only call (regenerated AI output varies between calls), which would have been valuable.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with what the tool produces and the modes, with the no-persistence guarantee prominent. There is minor redundancy between 'jamais enregistré' and 'rien n'est enregistré ni envoyé', but nothing that wastes real space.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists and the annotations are rich, so return values and safety need no explanation here. The remaining gap is input-side: with zero schema coverage and siren unexplained, the agent cannot confidently construct a call, and the description does not clarify the relationship between siren and terrain_id.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must carry parameter meaning, and it only partially does: the four mode values are enumerated and loosely glossed, but two of the three parameters — terrain_id and siren — are never mentioned. It is also unclear whether siren and terrain_id are alternatives or used together, a real ambiguity for the calling agent.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific capability: AI-generated proposals on a terrain — owner sentiment, draft agreement, summary — plus the modes that produce them. An agent can tell what the tool returns and that it is generative rather than a raw-data lookup. It stops short of naming a sibling it must not be confused with (e.g. toise_proprietaire for the underlying owner record).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Listing the four modes ('sentiment, brouillon, resume, courrier') implicitly tells the agent which capability to pick, and 'courrier' carries the note that the draft must be validated in the CRM. There is no explicit when-to-use vs when-not, no guidance on choosing this over a sibling such as toise_accords_fonciers for the actual agreement record, and no prerequisites.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources