Audit fraîcheur des CV / dossiers techniques

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

Beyond the annotations (readOnlyHint, idempotentHint), the description reveals a crucial behavioral trait: 'N'appelle aucune API BoondManager et ne lit aucune donnée : la réponse est générée côté serveur MCP.' It also specifies the exact output behavior: 'Returns : un runbook en texte — la liste ordonnée des appels Boond à effectuer ... rien n'est fait par cet appel.' This goes well beyond what the annotations convey and prevents the agent from expecting side effects or data reading.

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 longer than average but well-structured with clear labels ('Quand', 'Plutôt que', 'Returns') and front-loaded with the core purpose. Each section earns its place; the 'Plutôt que' detour is slightly verbose for a generic description but still highly relevant for tool selection. It is efficient enough for an agent to parse quickly.

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 workflow tool with no output schema and no required parameters, the description fully explains what the agent will receive ('un runbook en texte'), that no API calls or data reads happen, and that the model must execute the returned steps afterwards. This is complete and leaves no ambiguity about how to handle the tool's result.

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% since both manager_id and seuil_mois have descriptive text. The tool description does not add any additional parameter semantics — it merely repeats the workflow context. As a result, the baseline score of 3 is appropriate: the schema does the heavy lifting and the description neither improves nor detracts from it.

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: 'Identifie les ressources dont le CV ou le dossier technique est obsolète, incomplet, ou manquant' and adds a prioritization criterion ('Priorise celles bientôt sur le marché'). This clearly distinguishes it from the many sibling CRUD/search tools and even from other workflow_* tools by stating it produces a runbook rather than performing operations.

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 provides 'Quand : pour dérouler ce scénario multi-étapes sans avoir à retrouver soi-même le bon enchaînement d'outils et les bons noms de filtres' and gives an explicit alternative with conditions: 'Plutôt que : le prompt MCP `cvs_a_mettre_a_jour` si le client l'expose'. This tells the agent when to use the tool and when to prefer a cheaper alternative, which is exactly what an agent needs.

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