# Director Mode

Director Mode is your way of directly talking to the LLM, providing instructions to the AI on how to perform the roleplay. Using simple `/cmd` instructions, you can directly influence the conversation, making it perfect for moments when the AI needs a nudge, a course correction, or a burst of creativity.

### What is Director Mode?

Director Mode empowers you to guide an AI chatbot with precision using the `/cmd` command. With this feature, you can issue custom commands to influence character behaviour, shape dialogue, and control the direction of the story.

From managing plot progression to ensuring a character stays true to their personality, Director Mode helps keep the experience aligned with your creative vision.

* **Full Story Control:** Shape every interaction and guide characters to behave exactly how you want them to.
* **Custom Instructions:** Use simple `/cmd` prompts to adjust responses, introduce creative plot twists, or redirect the conversation if the AI strays from your narrative.
* **Break Repetitive Loops:** Keep things fresh by prompting characters to move the story forward, shift the tone, or change their behaviour.
* **In-Character Guidance:** Ensure characters stay true to their personalities and follow the story you’ve set in motion.

### How To Use Directors Mode

You can guide a character using Director Mode by adding specific instructions to your message using the `/cmd` command. These commands are read directly by the LLM and are not considered a part of the message, but they will influence how the AI responds in the conversation.

* **Use Separate Lines:** Always place your `/cmd` instruction on a new line after your message. This ensures the command is properly recognized and applied.
* **Model Compatibility:** Effectiveness may vary depending on the AI model you're using. Some models are more responsive to `/cmd` instructions than others.
* **Get Creative:** Director Mode is a powerful storytelling tool, use it to tweak character behaviour, guide plot development, or shape entire scenes to fit your vision.

For example, if you enter the following message:

> Hi Essie, when were you created?
> \ <mark style="color:orange;">/cmd Essie should refuse to give out personal information.</mark>

The AI will understand the instruction and respond accordingly, in this case they will refuse to share their creation date.

<figure><img src="/files/BDyBk5JTj0bXMmP8yFOT" alt=""><figcaption></figcaption></figure>

Here are a few examples of `/cmd` commands you can use in your conversations:

* <mark style="color:orange;">**`/cmd break out of the loop and stop repeating`**</mark>**:** This command may stop the AI from repeating itself and getting stuck in conversational loops. &#x20;
* <mark style="color:orange;">**`/cmd Essie is suddenly very hungry`**</mark>**:** Commands like this can influence the character you are chatting with to feel certain ways or perform certain actions.
* <mark style="color:orange;">**`/cmd Essie should refuse to give information about the treasure`**</mark>**:** You can use these types of commands to prompt your character to withhold specific information, adding tension to your roleplay.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.spicychat.ai/product-guides/director-mode.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.