pedir_ao_agente

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

With no annotations, the description carries the full burden and does disclose meaningful behavior: the Maestro turns a worker on and off around the call, the response is short, and the answer comes from a small local model whose facts must be verified. The sentinel return value 'PRECISA DO CLAUDE' is also a genuine behavioral disclosure that the agent must handle. It stops short of stating latency, cost, or failure modes, which would round out a dispatcher that can silently route work elsewhere.

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 routing purpose is front-loaded and the text is short, but the parenthetical '(pasta: projeto ...)' is jammed into the running prose in a way that reads like a stray parameter note rather than a structured line. The telegraphic, run-on phrasing costs a little parsing effort without adding information.

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?

An output schema exists, so return-shape explanation is not required, and the description still flags the one return value that changes agent behavior ('PRECISA DO CLAUDE'). For a two-parameter orchestrator this is close to complete; what is absent is any notion of the tool's cost or latency, which matters for something the description tells the agent to call on every turn.

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 0%, so the description must compensate and only partly does: 'Pedido em português' tells the agent the request must be written in Portuguese, and 'pasta: projeto (opcional, se o pedido não citar o nome)' explains that the folder is optional and only needed when the request does not name the project. Neither parameter gets format examples, naming conventions, or an explanation of what happens if the folder is omitted and unresolvable.

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 states a concrete routing behavior: the local Maestro picks which worker (Revisor, testes, git, Memória, or the Cérebro for simple questions) handles the request and returns a short answer. That is a real verb-plus-mechanism, not a restatement of the name. The gap is sibling differentiation: it never contrasts itself with perguntar_local or segunda_opiniao, which overlap heavily with the 'Cérebro para perguntas simples' path.

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?

It gives an unusually explicit trigger: 'Chame em toda conversa, antes de trabalhar' (call in every conversation, before working), plus a fallback rule when the sentinel 'PRECISA DO CLAUDE' comes back. What is missing is when NOT to use it, i.e. whether perguntar_local or segunda_opiniao should be preferred for direct local queries.

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