Alertes détectées sur le portefeuille surveillé : cessation SIRENE, nouvelle procédure collective BODACC, changement ou décès de dirigeant. C'est l'outil « quoi de neuf depuis mon dernier passage » — inutile de re-vérifier les SIREN un par un. marquer_lues=true acquitte les alertes retournées (le prochain appel ne rendra que les nouveautés). Clé API requise.
| Name | Required | Description | Default |
|---|---|---|---|
| api_key | Yes | ||
| marquer_lues | No | ||
| non_lues_seulement | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Despite annotations already indicating non-read-only behavior, the description adds meaningful detail: setting marquer_lues=true acknowledges alerts so subsequent calls return only new ones. It also discloses the API key requirement, providing useful operational context beyond the schema.
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 three sentences, front-loaded with the core purpose and alert types, followed by the stateful behavior and auth requirement. Every sentence earns its place with no redundancy or 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?
For a tool with no output schema and basic annotations, the description covers purpose, alert categories, stateful acknowledge behavior, and authentication. It lacks details about the return format and leaves non_lues_seulement under-explained, but it is sufficient for an agent to select and invoke the tool correctly.
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%, so the description must compensate. It clearly explains marquer_lues and mentions the api_key requirement, but it does not explain non_lues_seulement, which is a parameter with a default true value and non-obvious semantics. Partial coverage earns a middle score.
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 clearly identifies this as an alert-listing tool for a monitored portfolio, enumerating specific alert types (cessation SIRENE, BODACC, dirigeant changes). It positions itself as the 'what's new since my last visit' tool, which distinguishes it from sibling tools like resoudre_siren or verifier_siret.
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 provides strong usage context: use this to see new alerts rather than re-checking SIRENs individually. It also explains the stateful behavior of marquer_lues, though it does not explicitly state when not to use the tool or name alternative tools for other use cases.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have clearly distinct purposes (e.g., name-to-SIREN resolution vs SIRET verification), but croiser_deces and dirigeants_entreprise both involve death-registry matching, and the pair resoudre_siren/verifier_siret could initially confuse. However, the descriptions disambiguate well.
The naming mixes infinitive verbs (resoudre, verifier, croiser, diagnostiquer) with noun phrases (sante_entreprise, dirigeants_entreprise, surveillance_alertes), so there's no uniform verb_noun or noun pattern. Still, the names are reasonably readable and not chaotic.
With 8 tools, the server covers the core workflows of data validation, enrichment, scoring, and monitoring without bloat. Each tool addresses a distinct business need, and the count is well within the ideal range.
The surface covers identification, verification, scoring, and portfolio monitoring, but lacks a simple company detail retrieval endpoint (e.g., by SIREN) and a tool for updating/deleting non-portfolio data. These are minor gaps that can be worked around.
