Loading ToolBench
Servidor MCP para integração com a API TESS, permitindo utilizar agentes TESS via MCP
Server exposes 13 tools across multiple files (JS, TS, Python) with inconsistent quality. Tool naming is reasonable (verb_noun pattern mostly followed), but descriptions are sparse and generic. Critical gaps: most tools lack documented output schemas, parameter descriptions are minimal or missing, no error handling guidance, and no input validation rules. Tools 1-5 and 10-13 have basic descriptions but lack depth. Tools 6-9 (Portuguese names) duplicate functionality with slightly different signatures. No tool annotations (readOnlyHint/destructiveHint). Schema visibility is poor, parameters exist but lack type constraints, ranges, or format specifications. This server would require significant rework before production use.
Gera resposta para um prompt
Executa um agente com mensagens específicas
Verifica a saúde do servidor
Lista todos os agentes disponíveis na API TESS
Lista todos os arquivos associados a um agente
Obtém detalhes de um agente específico
Processa uma imagem e retorna informações
Duplicate tool definitions across languages (JS, TS, Python). Tools 1-5 and 6-9 perform identical operations with different names and signatures, forcing LLMs to reason about which variant to use. Consolidate to a single canonical set.
Output schemas not documented. Tools return data but no schema is visible in source, LLMs cannot plan downstream calls or extract required fields. Document return types for all 13 tools.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 40 | <=2025-11-25 | v2 |
Busca informações sobre um tópico
Executa um agente específico no TESS.
Obtém detalhes de um agente específico no TESS.
Obtém o status atual de uma execução.
Lista os agentes disponíveis no TESS.
Faz upload de um arquivo para o TESS.
Parameter descriptions are minimal or missing. 'temperature' described as 'Temperatura para geração (de 0 a 1)' lacks actionable detail (what does 0 vs 1 mean? Is it required?). 'query' in search_info has no description at all. Add 50-150 char descriptions explaining purpose, constraints, and examples.
No error handling guidance. Tools throw generic errors ('Erro ao listar agentes') without recovery hints. LLMs cannot determine if errors are retryable, user-fixable, or fatal. Add error classification and actionable recovery steps.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). LLMs cannot distinguish safe reads from destructive writes. Mark tess.execute_agent and tess.upload_file as destructive; mark list/get tools as readOnly.