compute_tvaCalcule la déclaration de TVA pour les biens para-hôteliers assujettis (régime TVA liable) sur une année donnée. Retourne la TVA collectée, déductible et le solde à payer ou à récupérer, ventilés par bien, par taux et par trimestre. Tous les montants sont en euros.
| Name | Required | Description | Default |
|---|---|---|---|
| year | Yes |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The readOnlyHint annotation already indicates this is a read-only operation, so the description adds value by detailing what is returned (TVA collectée, déductible, solde) and the aggregation dimensions (bien, taux, trimestre). This goes beyond the annotation and clarifies the calculation scope, though it does not mention error handling or prerequisites.
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 exceptionally concise: two sentences that front-load the main verb and purpose, then detail outputs. No redundant information or filler. Every phrase adds meaning, from the target population to the currency.
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 tool has only one parameter and no output schema, the description adequately covers the return value structure and scope. It explains the outputs and their segmentation, which is sufficient for an agent to understand what the tool provides. It misses only minor details like data source dependencies (existing properties, fiscal year definitions), but these are not critical for effective invocation.
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 only defines 'year' as an integer with no description (0% coverage), so the description must compensate. It does so by stating 'sur une année donnée' (for a given year), directly clarifying the parameter's meaning. However, it does not specify constraints like valid year range or format beyond the implicit integer type, leaving some ambiguity for edge cases.
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 a specific verb ('Calcule' / calculates) and resource (VAT declaration for taxable para-hotel properties) along with the output composition (collected VAT, deductible VAT, balance) and breakdown (by property, rate, quarter). This distinguishes it from sibling tools like compute_depreciation or generate_tax_return, which address different financial scopes.
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 usage by specifying the context (para-hotel properties, taxable regime, given year) but does not explicitly contrast with alternatives such as generate_tax_return or compute_fiscal_year. There is no 'when-not-to-use' or mention of sibling tools, leaving the agent to infer applicability rather than receiving explicit selection guidance.
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.
The set contains two almost identical comparison tools: compare_micro_bic and get_simulation both compute micro-BIC vs réel and recommend the best regime. This creates a direct selection conflict. Other tools are mostly distinct, but this overlap is significant.
Tool names follow a verb_noun pattern consistently, with list_properties, create_income, delete_expense, etc. Minor deviations exist such as compare_micro_bic (no standard noun) and the plural/singular mismatch between delete_document and export_documents, but these are easily understood.
With 44 tools, the surface is very large for an MCP server. While the LMNP domain is broad, the number exceeds the typical range and will be hard for agents to navigate efficiently. Some functions could be consolidated (e.g., get_simulation absorbs compare_micro_bic).
The set covers CRUD for properties, incomes, expenses, furniture, and works, plus documents and fiscal computations. However, there is no create/update/delete for loans (only compute and get), no delete_property, and no update for property components. These gaps could block common workflows.