---
title: "Bot settings reference"
description: "Every setting on your bot's General settings tab - what it does, its default and range, and which plans include it."
canonical_url: "https://chatthing.ai/docs/bot-settings/settings-reference"
last_updated: "2026-09-25"
---

# Bot settings reference

All of these settings are on your bot's **General settings** tab, which team owners and admins can open. Change a setting, then click **Update bot** at the bottom of the page to save.

Most settings can also be changed from an AI assistant with the [`update_bot` MCP tool](https://chatthing.ai/docs/mcp/tools). The field names match the setting names in the tables below.

## General settings

| Setting | What it does | Default | Plans |
| --- | --- | --- | --- |
| **Bot name** | The bot's name. | - | All |
| **Bot description (markdown supported)** | Shown only on the bot's landing page. It is **not** sent to the AI model: put instructions in **Prompt**. | Empty | All |
| **Welcome message** | The first message people see before they type anything. Required. | - | All |
| **Example questions** | Suggested questions people can click to start a chat. Up to 10. | None | All |
| **Prompt** | The bot's instructions: its role, tone, rules and what to do when it can't help. Blank uses the default prompt. **Generate with AI** writes one for you. See [write your bot's instructions](https://chatthing.ai/docs/bot-settings/instructions). | Default prompt | All |
| **Allow users to provide feedback on responses** | Shows thumbs up and thumbs down buttons on each reply in the chat widget. Ratings appear in [conversations](https://chatthing.ai/docs/manage/conversations), where you can filter by them. | Off | All |
| **Include sources for responses** | Shows links to the web pages an answer came from under each reply in the chat widget. Only website data sources are shown. | Off | All |
| **Allow human takeover of conversations** | Lets your team claim a live website chat and reply as themselves. Only for chats started in the website widget. Works best with the [Talk to a human](https://chatthing.ai/docs/power-ups/talk-to-a-human) power-up. See [human takeover](https://chatthing.ai/docs/manage/human-takeover). | Off | Standard, Pro, Enterprise |

Older bots show **System message**, **Use our default prompt** and **Initial prompt** instead of a single **Prompt**. See [migrating to a single prompt](https://chatthing.ai/docs/bot-settings/instructions#i-see-system-message-and-initial-prompt-instead-of-prompt).

## Model settings (advanced)

We recommend leaving these at their defaults until you've tested your bot.

| Setting | What it does | Default | Range | Plans |
| --- | --- | --- | --- | --- |
| **Model** | The AI model that writes replies. See [models](https://chatthing.ai/docs/bot-settings/models). | Default model | - | Standard, Pro, Enterprise (others use the default) |
| **Enable file upload (multi-modal)** | Lets people attach images in the chat widget: PNG, JPG and WebP, up to 20 MB each. Only shown for models that support vision. | Off | On / off | All |
| **Enhanced knowledge retrieval** | Rewrites follow-up questions into standalone questions before searching your knowledge. Uses more tokens. See [retrieval settings](https://chatthing.ai/docs/knowledge/retrieval-settings). | Off | On / off | All |
| **Creativity (temperature)** | How varied replies are. Lower values give consistent, focused answers that stick to your content and prompt; higher values give more varied answers but make the bot more likely to fill gaps or make things up. | 0.3 | 0–1 | All |
| **Document relevance** | How closely content must match a message before the bot is given it. See [retrieval settings](https://chatthing.ai/docs/knowledge/retrieval-settings). | 0.75 | 0.25–0.9 | All |
| **Document character limit** | The size of each piece your content is split into when it syncs. See [retrieval settings](https://chatthing.ai/docs/knowledge/retrieval-settings). | 2,000 | 1,000–4,000 | All |
| **Max context size (base tokens)** | The token budget for each reply's prompt, knowledge and conversation history. Higher can improve answers but uses more tokens. | Set by the model | 1,024 to the model's maximum | All |
| **Max context documents** | How many pieces of your content the bot is given per message. `0` turns knowledge retrieval off. See [retrieval settings](https://chatthing.ai/docs/knowledge/retrieval-settings). | 4 or 6 | 0–12 | All |
| **Max message history** | How many earlier messages in the conversation the bot sees. Lower uses fewer tokens but the bot remembers less. | 12 | 0–50 | All |

## Usage limits

Limits that stop a bot using too many tokens or being abused. See [usage limits](https://chatthing.ai/docs/bot-settings/usage-limits) for what people see when a limit is reached.

| Setting | What it does | Default |
| --- | --- | --- |
| **Daily token limit** | Stops the bot replying once it has used this many message tokens in a day (UTC). | Off, 1,000,000 |
| **Set an email notification threshold** | Emails team owners and admins when the bot's daily usage passes this number. Must be lower than the daily limit. | 750,000 |
| **Maximum message length** / **Maximum characters per message** | Rejects any single message longer than this. | On, 6,000 characters for new bots |

## Transfer ownership

Moves the bot to another team. Only shown if you belong to more than one team. You need to be a member of the new team with permission to create bots, and the team needs space for another bot on its plan. See [teams](https://chatthing.ai/docs/account/teams).

## Delete bot

**Delete bot** at the bottom of the page permanently removes the bot and all of its data. This can't be undone. To keep a copy first, [duplicate the bot](https://chatthing.ai/docs/bot-settings/duplicate-a-bot).

## Related

- [Write your bot's instructions](https://chatthing.ai/docs/bot-settings/instructions): Write the prompt that sets your bot's role, topics, tone, answer length, language, power-up use and handoff, with before-and-after examples.
- [Retrieval settings](https://chatthing.ai/docs/knowledge/retrieval-settings): Control how much of your knowledge your bot reads for each message - document relevance, context documents, enhanced retrieval, sources and chunk size.
- [Limit how many tokens your bot can use](https://chatthing.ai/docs/bot-settings/usage-limits): Cap a bot's daily message token use, get an email before it hits the cap, limit message length, and see what people see when a limit is reached.