Agents use grammar (#31735)

* Allow optional use of grammars to constrain generation

docs/source/en/agents.md

@@ -119,10 +119,12 @@ def llm_engine(messages, stop_sequences=["Task"]) -> str:
 ```
 
 You could use any `llm_engine` method as long as:
-1. it follows the [messages format](./chat_templating.md) for its input (`List[Dict[str, str]]`) and returns a `str`
-2. it stops generating outputs at the sequences passed in the argument `stop`
+1. it follows the [messages format](./chat_templating.md) (`List[Dict[str, str]]`) for its input `messages`, and it returns a `str`.
+2. it stops generating outputs at the sequences passed in the argument `stop_sequences`
 
-You also need a `tools` argument which accepts a list of `Tools`. You can provide an empty list for `tools`, but use the default toolbox with the optional argument `add_base_tools=True`.
+Additionally, `llm_engine` can also take a `grammar` argument. In the case where you specify a `grammar` upon agent initialization, this argument will be passed to the calls to llm_engine, with the `grammar` that you defined upon initialization, to allow [constrained generation](https://huggingface.co/docs/text-generation-inference/conceptual/guidance) in order to force properly-formatted agent outputs.
+
+You will also need a `tools` argument which accepts a list of `Tools` - it can be an empty list. You can also add the default toolbox on top of your `tools` list by defining the optional argument `add_base_tools=True`.
 
 Now you can create an agent, like [`CodeAgent`], and run it. For convenience, we also provide the [`HfEngine`] class that uses `huggingface_hub.InferenceClient` under the hood.