Loja Aurora
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Servidor de dados da Loja Aurora (e-commerce fictício). Todas as ferramentas são somente leitura. Valores em reais (BRL). Faturamento e vendas desconsideram pedidos cancelados. Prefira as ferramentas específicas; use run_sql apenas quando nenhuma delas responder a pergunta.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_orderB | Detalhes de um pedido pelo número: status, data, cliente, itens e valor total. |
| find_customerA | Busca clientes por parte do nome, e-mail ou ID. Retorna cidade, nº de pedidos e total gasto. |
| customer_ordersA | Lista os pedidos mais recentes de um cliente (use find_customer antes para obter o ID). |
| sales_summaryA | Faturamento, nº de pedidos, itens e ticket médio num período (datas AAAA-MM-DD, inclusivas). group_by: 'total', 'month' (por mês), 'category' (por categoria) ou 'state' (por UF). |
| top_productsB | Produtos mais vendidos num período (datas AAAA-MM-DD), por faturamento ou quantidade. |
| describe_schemaA | Mostra as tabelas e colunas do banco. Use antes de escrever uma consulta em run_sql. |
| run_sqlA | Executa UMA consulta SQL SELECT (SQLite) somente leitura, até 50 linhas. Use só quando nenhuma outra ferramenta responder. Escritas são bloqueadas. |
| search_policiesB | Busca nas políticas da loja: trocas, devoluções, frete, prazos, garantia, pagamento e atendimento. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Each tool targets a distinct resource: customer lookup, orders per customer, order detail by number, sales aggregates, top products, policy text, plus a schema and SQL pair. The only fuzziness is between customer_orders/get_order and sales_summary/run_sql, but descriptions make the intended precedence clear (run_sql is explicitly a last resort).
All names are snake_case and readable, but the conventions are mixed: verb_noun (find_customer, get_order, search_policies, run_sql, describe_schema) alongside noun-first/aggregate names (customer_orders, sales_summary, top_products). Predictable enough to use, but not a single consistent pattern.
Eight tools is well-scoped for a small store analytics/assistant domain. Each tool earns its place, and the run_sql/describe_schema pair is a sensible controlled escape hatch rather than padding.
Covers the main read paths: customer lookup, order history, order detail, period sales, top products, and policy search, plus SQL fallback for anything else. Minor gaps exist for product catalog/stock lookup and structured return or exchange records, but the SQL escape hatch lets agents work around most of them.