Recherche dans les guides pratiques de constat-auto.com : constat amiable, démarches après un accident, responsabilité, indemnisation, assurance auto en France. Renvoie une liste JSON de résultats avec identifiant, titre, URL et extrait. Utilisez ensuite fetch avec l'identifiant pour lire un guide en entier.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | Termes de recherche en langage naturel. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare read-only, idempotent, and non-destructive behavior. The description adds valuable context beyond this by detailing the return fields (identifier, title, URL, excerpt) and the follow-up action with fetch, which is especially helpful given no output schema exists.
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 concise, consisting of two well-structured sentences that immediately convey the purpose, then the output format, and finally the next step. No unnecessary words or repetition.
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?
Given the simple tool with one parameter and no output schema, the description fully covers the search scope, return format, and how to proceed. It's complete and actionable for an agent.
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 schema fully describes the only parameter (query) with a natural language description, so the description doesn't need to add more. It adds no extra semantics beyond what the schema provides, making the baseline score of 3 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?
The description clearly states the tool searches practical guides on constat-auto.com with specific topics, and explicitly mentions the JSON result format with identifier, title, URL, and excerpt. It distinguishes from siblings by directing users to use `fetch` with the identifier for reading full guides, making its role unique.
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 implies its use for finding guides and explicitly instructs to use `fetch` next, providing a clear workflow. However, it doesn't explicitly state when not to use this tool or mention alternatives like the constat_* tools, though the context makes it implicit.
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.
Each constat_* tool corresponds to a distinct phase of the accident statement workflow—starting, filling, sketching, signing, status checking, finalizing, canceling—with no two tools overlapping in purpose. The two guide-related tools (search, fetch) are clearly separate from the workflow tools.
The constat_ prefix provides a recognizable pattern, but the suffix alternates between verbs (demarrer, renseigner, finaliser, annuler) and nouns (circonstances, croquis, signatures, statut, assureurs). Additionally, the two guide tools (fetch, search) break the pattern entirely, mixing languages and conventions.
With 11 tools, the server is well-scoped for its purpose: each tool serves a clear step in the constat process and the two guide tools complement the primary workflow without redundancy. The count is within the ideal 3–15 range and does not feel heavy or thin.
The workflow is well covered: start, fill, sketch, sign, monitor status, finalize, and cancel. A minor gap is the lack of a dedicated 'get full constat contents' tool (statut only shows missing fields and signature state), but the core lifecycle is otherwise complete.
