# Add existing voice
Source: https://docs.brivvy.io/add-existing-voice

Create a voice profile by uploading real examples of your brand in action.

Add existing voice lets you skip manual configuration. Instead of setting tone sliders and rules one at a time, you provide examples of your writing and Brivvy extracts a voice profile automatically.

This is the fastest way to get started if your brand already has published content, style guides or marketing copy that represents how you communicate.

## Before you start

* Any member of your workspace can create a voice. There are no role restrictions.
* Prepare at least one example of your brand writing. The more representative your examples, the more accurate the extracted voice.

## How to add an existing voice

1. In your Brivvy workspace, navigate to the **Voices** tab.
2. Select **Upload existing**. If you already have voices in your workspace, open the dropdown on the **New brand voice** button and select **Upload existing**.
3. In the **Type of voice** dropdown, choose a category that fits your use case. This is optional, but it helps Brivvy name and contextualize the extracted voice.
4. Under **Examples**, add your brand materials using one or more of the three source types: [URL](#url), [Text](#text) or [File](#file). You can mix source types in a single voice.
5. Select **Continue**.
6. Wait for extraction to complete. This typically takes a couple of minutes. Brivvy analyzes your examples and maps patterns to tone dimensions and writing rules.
7. Review the extracted voice configuration. Check that the tone sliders and rules reflect how your brand actually communicates. Adjust anything that does not look right.
8. Select **Publish** to make the voice available across your workspace.

## Source types

You can provide examples through three input methods. Each one feeds the same extraction engine, so choose whichever format is most convenient for the content you have.

### URL

Paste a link to a published page that represents your brand voice. Brivvy reads the content of that single page and uses it as an extraction input.

<Note>
  URLs are scoped to a single page. Brivvy does not crawl an entire domain. This is intentional. The goal is quality over quantity, so choose pages that best represent how your brand communicates.
</Note>

### Text

Paste raw text directly into the wizard. This works well for content that is not published online, such as internal messaging guidelines, draft copy or email templates. Each text sample must be at least 100 characters.

### File

Upload a document that contains examples of your brand writing. This is useful for style guides, brand books or long-form content stored locally.

## Tips for better results

* **Choose standout examples.** The wizard works best when you provide content that clearly represents your intended voice. A polished blog post or a well-written landing page is more useful than a generic status update.
* **Mix your sources.** Combining a URL, a text snippet and a file gives Brivvy a broader picture of how your brand communicates across formats.
* **Provide two to five examples.** A single example can work, but multiple inputs help Brivvy identify consistent patterns rather than one-off stylistic choices.
* **Review before publishing.** Extraction is a starting point. Always check the generated tone sliders and rules against your expectations before publishing.

## How to know it worked

After publishing, open the voice and confirm that the tone dimensions and rules match your brand. You can also use [Preview brand voice](/preview-brand-voice) to see how the configuration shapes AI output in practice.

## What happens if extraction fails

In rare cases, extraction may not complete successfully. If this happens, Brivvy notifies you and you can try again. Common fixes include providing longer or more varied examples and making sure URLs point to pages with substantial text content.

## What's next

* [Preview brand voice](/preview-brand-voice) to test the extracted configuration before using it in production.
* [Create a voice from scratch](/start-from-scratch) if you prefer to set tone and rules manually.
* [Connect an AI tool](/mcp-server) to start generating content with your new voice.


# Avoid Terms
Source: https://docs.brivvy.io/avoid-terms

Flag words and phrases your team should not use and define the alternatives that should appear instead.

Avoid terms are words and phrases your organization has decided not to use. When you add an avoid term to the [glossary](/glossary-overview), AI tools connected through the MCP server will never produce that term during content generation. Instead, they substitute one of the alternatives you specify. This is how you eliminate jargon, outdated language and filler phrases without relying on manual editing.

## When to use avoid terms

Add an avoid term when a word or phrase consistently appears in your content but does not meet your brand's standards. Common examples include corporate jargon ("leverage," "utilize"), filler phrases ("in order to") and words that sound vague or overused ("delve"). Avoid terms are also useful for retiring outdated product names or deprecated terminology that team members might still reach for out of habit.

## Fields

Each avoid term supports the following fields.

**Name.** The word or phrase to block. AI tools will never produce this term in generated content. Enter it exactly as it appears in typical usage, including capitalization. If the term can appear in multiple forms, add each form as a separate entry.

**Related names.** A list of one or more recommended alternatives. When AI tools encounter a context where the avoided term would naturally appear, they use one of these alternatives instead. The tool selects the alternative that best fits the surrounding sentence. You should provide at least one related name for every avoid term so the tool has a clear replacement path.

## Examples

Here are a few examples to illustrate how avoid terms work in practice.

**Corporate jargon.** Name: "utilize." Related names: "use," "apply." Whenever content would include "utilize," the AI tool substitutes "use" or "apply" depending on context.

**Filler phrase.** Name: "in order to." Related names: "to." This removes a three-word phrase that adds no meaning and replaces it with a single word.

**Overused verb.** Name: "leverage." Related names: "use." A straightforward swap that keeps content clear and direct.

**Vague language.** Name: "delve." Related names: "explore," "dig into," "look at." Multiple alternatives give the AI tool flexibility to choose the option that reads most naturally in each sentence.

## How alternatives are selected

When an avoid term has multiple related names, the AI tool evaluates the surrounding context and picks the alternative that fits best. You do not need to specify when to use which alternative. However, ordering matters as a soft signal. The first related name in the list is treated as the default choice, and the tool will prefer it when multiple alternatives are equally appropriate.

## Best practices

* Always provide at least one alternative. An avoid term without a related name gives the AI tool no replacement path, which can produce awkward phrasing.
* Be specific about what you are blocking. "Leverage" as a verb and "leverage" as a noun may need separate entries if you only want to block one usage.
* Do not overload the avoid list. A focused list of 10 to 20 terms is more effective than a sprawling list of 100. Target the terms that actually appear in your content.
* Review avoided terms alongside your [preferred terms](/preferred-terms). Sometimes a term belongs in the preferred list rather than the avoid list, especially if the issue is inconsistent spelling rather than unwanted language.
* Check your list against real content. If your team never actually writes "delve," you do not need an entry for it. Focus on the terms that cause real problems.

## Related

* [Glossary](/glossary-overview) for an overview of how the glossary works.
* [Preferred terms](/preferred-terms) to define the correct names and acronyms your team should always use.
* [Voices](/voices) to configure tone and rules that work alongside your terminology.


# Billings and plans
Source: https://docs.brivvy.io/billings-plans





# ChatGPT
Source: https://docs.brivvy.io/chatgpt

Connect Brivvy to ChatGPT so the model can access brand voice rules and generate on-brand content directly in chat.

ChatGPT supports remote MCP servers through its Connectors feature, which means you can add Brivvy without editing any config files. Once connected, ChatGPT can call Brivvy tools mid-conversation to retrieve brand voice guidelines and list templates.

Brivvy's MCP server uses tools beyond the default `search` and `fetch` pattern, so Developer Mode must be enabled for the integration to work. Without it, ChatGPT will reject the connection.

<Note>
  MCP connectors and Developer Mode require a ChatGPT Pro, Plus, Business, Enterprise or Edu plan.
</Note>

## Before you start

* A Brivvy account on a plan that includes MCP access.
* A ChatGPT Pro, Plus, Business, Enterprise or Edu account.
* For team workspaces: admin access to enable Developer Mode at the workspace level.

<iframe title="YouTube video player" />

<Callout icon="youtube">
  If you prefer video tutorials, check out our [YouTube](https://www.youtube.com/@brivvyHQ) page.
</Callout>

## Setup

<Tabs>
  <Tab title="For admins">
    1. In ChatGPT, click your workspace name in the bottom left sidebar and select **Workspace settings**.
    2. Navigate to **Apps**.
    3. Click **Create**, and select **Enable developer mode**.
    4. Fill in the required fields:
       | <br /> **Field** | **Value**                                     |
       | :--------------- | :-------------------------------------------- |
       | Connector name   | Brivvy                                        |
       | Description      | Access Brivvy brand voice rules and templates |
       | MCP server URL   | `https://mcp.brivvy.io`                       |
    5. Agree to terms and click **Create**.
    6. Connect Brivvy with ChatGPT via **Authorization process**.
    7. **Publish** the Brivvy app to your workspace.
  </Tab>

  <Tab title="For members">
    Once your workspace admin has enabled Developer Mode (or if you are on an individual Pro or Plus plan), follow these steps to connect Brivvy.

    1. In ChatGPT, click your profile picture in the top right and go to **Settings > Apps > Advanced settings**. Toggle on **Developer Mode**.
    2. Go to **Settings > Connectors** and click **Create**.
    3. Fill in the connector details:
       | Field          | Value                                         |
       | -------------- | --------------------------------------------- |
       | Connector name | Brivvy                                        |
       | Description    | Access Brivvy brand voice rules and templates |
       | MCP server URL | `https://mcp.brivvy.io`                       |
    4. Make sure **Authentication** is set to **OAuth** and click **Create**.
    5. ChatGPT will open an authorization page. Review the requested permissions and click **Authorize** to grant ChatGPT access to Brivvy on your behalf.
    6. Start a new chat, click the tools menu near the message input and select **Developer Mode**. Choose **Brivvy** from the list. The connector stays active for the entire conversation once enabled.
  </Tab>
</Tabs>

## How to know it worked

Send a prompt that requires brand voice context, for example: "Using Brivvy, get the brand voice rules and write a two-sentence product update." ChatGPT will call the appropriate Brivvy tool and return a response shaped by your brand voice rules. Tool-call payloads are displayed in the UI as ChatGPT works, showing inputs and outputs for each Brivvy tool called.


# Claude
Source: https://docs.brivvy.io/claude

Connect the Brivvy MCP Server with Claude's web interface and desktop application to access brand voice during content generation.

Claude offers the most seamless integration with Brivvy's MCP Server across multiple environments. Teams using Claude for content generation can access brand voice guidelines, template instructions and glossary terms without leaving their workflow. The integration works identically in Claude's web interface and desktop application, ensuring consistent brand application regardless of where content is created.

Claude's web interface and desktop application share the same MCP configuration. Settings configured in one location apply to both environments.

<iframe title="YouTube video player" />

<Callout icon="youtube">
  If you prefer video tutorials, check out our [YouTube](https://www.youtube.com/@brivvyHQ) page.
</Callout>

## Setup

<Tabs>
  <Tab title="For admins">
    Before your team can use Brivvy, a Claude admin must add it as a custom connector from Organization settings. This makes the connector available for team members to connect their own accounts.

    1. In Claude, click your profile or workspace name and navigate to **Organization settings > Connectors**.
    2. Scroll to the bottom of the Connectors section and click **Add custom connector**.
    3. Fill in the following fields:
       | Field          | Value                                          |
       | -------------- | ---------------------------------------------- |
       | Name           | Brivvy                                         |
       | MCP server URL | [https://mcp.brivvy.io](https://mcp.brivvy.io) |
    4. Click **Add** to save. The Brivvy connector will now appear in your organization's connector list, ready for team members to connect their own accounts.
    5. Navigate to your individual **Settings > Connectors**, find Brivvy in the list and click **Connect**. Complete the OAuth authentication flow to authorize access for your own account.
  </Tab>

  <Tab title="For members">
    Once your organization's admin has added the Brivvy connector, each team member needs to connect their own account before using it.

    1. In Claude, click your profile name and navigate to **Settings > Connectors**.
    2. Locate Brivvy in the list of available connectors and click **Connect**.
    3. Your browser will open an authorization page. Review the requested permissions and click **Authorize** to grant Claude access to Brivvy on your behalf.
    4. After connecting, click **Configure** next to Brivvy in **Settings > Connectors** to authorize all tools at once. This means Claude won't prompt you for approval each time it uses a Brivvy tool.
    5. In any Claude conversation, click the **+** button at the bottom left of the chat, then select **Connectors**. Toggle Brivvy on to enable it for that conversation.
  </Tab>
</Tabs>

## How to know it worked

Send a prompt that requires brand voice context, for example: "Using Brivvy, get the brand voice rules and write a two-sentence product update." Claude will call the appropriate Brivvy tool and return a response shaped by your brand voice rules rather than a generic output.


# Claude Code
Source: https://docs.brivvy.io/claude-code

Connect the Brivvy MCP Server with Claude Code to access brand voice and templates from the command line.

Claude Code supports remote Model Context Protocol (MCP) servers over the Streamable HTTP transport with built-in OAuth authentication. You can add Brivvy with a single command in your terminal. Once connected, Claude Code can call Brivvy tools to retrieve brand voice rules, glossary terms and template instructions without leaving your shell. This works well for generating on-brand README files, release notes, commit messages and user-facing copy while you code.

## Before you start

* Claude Code installed and signed in to your Anthropic account.
* A Brivvy account.
* Terminal access with permission to run `claude` commands.

## Steps

1. **Add Brivvy as a remote MCP server.** Open your terminal and run the following to register the server:

```bash theme={null}
   claude mcp add --transport http brivvy https://mcp.brivvy.io
```

By default, Claude Code uses local scope, so the server is only available in the current project. To make Brivvy available across all your projects, add the `--scope user` flag:

```bash theme={null}
   claude mcp add --scope user --transport http brivvy https://mcp.brivvy.io
```

2. **Start a new Claude Code session.** Run `claude` in your terminal. Configuration changes take effect when a new session starts, so any existing sessions will not see Brivvy until you restart them.
3. **Open the MCP management menu.** Inside Claude Code, type `/mcp` to list configured servers. You should see `brivvy` with a "needs authentication" status.
4. **Authenticate with Brivvy.** Select `brivvy` from the menu, then select **Authenticate**. Your browser will open an authorization page where you can sign in to Brivvy and approve access. After you authorize, Claude Code stores the OAuth credentials locally.
5. **Confirm the connection.** Back in the terminal, you should see "Authentication successful". Run `/mcp` again and confirm Brivvy shows a connected status. Expand the entry to see the available tools: `list_voices`, `get_voice`, `list_templates` and `get_template`.

## How to know it worked

Send a prompt that requires brand voice context, for example: "Using Brivvy, get the brand voice rules and write a two-sentence release note for the latest version." Claude Code will call the appropriate Brivvy tool and return a response shaped by your brand voice rules rather than a generic output.


# Brivvy Concepts
Source: https://docs.brivvy.io/concepts

Gain a basic understanding of Brivvy's design so teams can set up their workspace and navigate efficiently.

## Workspace

A [workspace](/workspaces) holds all your templates, voices and other data for you and your team. You can be a member of multiple workspaces, but each has its own subscription and data.

<img alt="Workspaces 3" />

## Members

[Members](/members-roles) are the users in your workspace who can view and take actions in Brivvy. Brivvy offers two types of roles for members, Admins and Members.

<img alt="Members" />

## Voices

[Voices](/voices) define how your team sounds. You can create multiple voices to serve different use cases, such as a technical voice for documentation, a warm voice for marketing or a concise voice for social media. AI tools connected through the MCP server use your voices to generate on-brand content.

<img alt="Voices" />

## Templates

A [template](/create-template) is a reusable structure for a specific content type. It defines format and organization, not voice. Your team can use templates to standardize blog posts, product updates, documentation and more.

<img alt="Templates" />

## Labels

A label is a tag you apply to templates and other items in your workspace. Use labels to organize content by theme, audience, department or any category that fits your workflow.

<img alt="Labels" />

## MCP Server

The [MCP server](/mcp-server) connects your Brivvy workspace to AI tools like Claude, ChatGPT and Cursor. Once connected, these tools can access your voices and templates to generate on-brand content without leaving your workflow.


# Create comm
Source: https://docs.brivvy.io/create-comm

Comms are the primary way to manage your formal communications in Brivvy.

<Frame>
  <img title="Create Comms Cover Light" alt="Create Comms Cover Dark" />

  <img title="Create Comms Cover Light" alt="Create Comms Cover Light" />
</Frame>

## Overview

Comms are formal communications designed for structured, thoughtful content rather than quick messages. Unlike instant messaging in Slack, comms are intended for content that requires more consideration and typically has a broader or more permanent audience.

Comms are commonly used for:

* Product announcements
* Changelogs
* Blog posts
* Job postings
* Quarterly updates
* Client-facing communications

## Creating a comm

Brivvy works where you work. Whether you prefer to create comms inside Brivvy or outside is entirely up to you.

#### Creating with Brivvy

Navigate to the Comms page and click the **New Comm** button to open the comm creation interface.

#### Creating with chat client

You'll soon be able to create comms directly from AI clients like Claude or ChatGPT via the MCP server. When creating content in these AI clients, you can instruct them to create a comm in Brivvy, or the AI client will prompt you to save your work as a comm.

## Comm properties

Comms support the following properties:

| Property    | What it means                                                                       |
| ----------- | ----------------------------------------------------------------------------------- |
| Name        | A descriptive title for your comm                                                   |
| Status      | The current state of your comm                                                      |
| Content     | The main content of your comm                                                       |
| Assignee    | The person responsible for the comm                                                 |
| Template    | If the comm was created using a template, this indicates which template was applied |
| Brand voice | The brand voice used to generate the comm                                           |
| Creator     | The person who created the comm                                                     |
| Date        | Timestamp information for the comm                                                  |
| Source      | Where the comm was originally created, such as Claude or ChatGPT                    |

When creating a comm, you can apply a template to structure your content. For example, you might create a changelog comm using your changelog template, or a blog post comm using your blog template.

## Best practices

Since comms are designed for more formal, public-facing, or permanent communications:

* Take time to review and refine your content before publishing
* Use appropriate templates to maintain consistency across similar comm types
* Assign comms to the appropriate team member for accountability
* Keep comm status updated to reflect the current stage of your communication


# Templates Overview
Source: https://docs.brivvy.io/create-template

Learn how to create a content template in Brivvy to standardize posts with categories, inspiration sources, and AI-ready prompts.

## Overview

Templates are structured frameworks for creating consistent content in Brivvy. Use them to standardize blog posts, product updates, explainers, and more.

Templates define structure and formatting, not voice. Keep those parameteres strictly in your Brand Voice.

Templates consist of five core fields: Name, Description, Category, Inspired by and Prompt. The prompt field contains the complete instructions that content creators follow when generating new posts.

## When to use templates

Use a template when you need repeatable quality across writers, teams, or topics. Templates work best when you already know the intended content type and audience.

If you are unsure what to create, start in Discover to review existing templates and categories.

## Template fields

Every template requires five specific fields that define its purpose and usage.

* **Name**: A distinctive identifier for the template. Names should be memorable and avoid conflicts with existing categories or templates.
* **Description**: A 30-60 word explanation following the pattern: style → use case → unique trait → user benefit. This is what people scan in Discover.
* **Category**: The content type classification. Templates must be assigned to one of the 22 available categories. Browse and validate categories in Discover.
* **Prompt**: Complete instructions for content creation, including structure, guidelines and formatting requirements. This is what you run with Brivvy AI.

## How to create a template in Brivvy

<Steps>
  <Step title="Name the template">
    Pick a distinctive name that does not match a category name.

    <Tip>
      For example, the "Listicle"
    </Tip>
  </Step>

  <Step title="Write the description">
    Template descriptions follow a specific format to ensure consistency and clarity. Each description must be 30-60 words and flow naturally through four key elements.

    The description begins with style, establishing the content's tone and approach. For example: "A decision-driven Buyer's Guide format..."

    Next comes use case, explaining when content creators should apply this template. This might reference specific scenarios, topics or content goals.

    The unique trait distinguishes this template from similar options. What makes this approach different or valuable compared to other templates in the same category?

    Finally, the description articulates user benefit. How does this template help content creators or readers achieve their goals?
  </Step>

  <Step title="Select category">
    Pick the primary content type.
  </Step>

  <Step title="Build the prompt">
    Write the full instructions the writer (or Brivvy AI) will follow.
  </Step>

  <Step title="Publish prompts" />
</Steps>


# Cursor
Source: https://docs.brivvy.io/cursor

Connect the Brivvy MCP Server with Cursor to access brand voice and templates during AI-assisted coding and content generation.

Cursor supports remote MCP servers over Streamable HTTP with built-in OAuth authentication. You can add Brivvy by dropping a single entry into your MCP configuration file. Once connected, Cursor's Agent can call Brivvy tools to retrieve brand voice rules and template instructions directly in your coding workflow. This is useful for generating on-brand README files, API documentation, code comments and user-facing text without leaving the editor.

## Before you start

* Cursor installed (v1.0 or later).
* A Brivvy account on a plan that includes MCP access.

## Steps

1. Open Cursor and click the gear icon in the top right to open **Settings**. In the left sidebar, click **Tools & Integrations**.

2. Under **MCP Tools**, click **Add Custom MCP**. Cursor opens the `mcp.json` configuration file in the editor.

3. Paste the following configuration. If an `mcpServers` block already exists, add `"brivvy"` as an additional entry rather than creating a second block.

   ```json theme={null}
   {
     "mcpServers": {
       "brivvy": {
         "url": "https://mcp.brivvy.io"
       }
     }
   }
   ```

4. Save the file. Brivvy will appear in your MCP Tools list with a **Needs login** status.

5. Click **Needs login** next to Brivvy. Cursor opens a browser window where you can sign in to Brivvy and authorize the connection. After you approve, Cursor stores the OAuth credentials locally.

6. Back in Cursor, confirm the Brivvy server shows a green **Connected** status under MCP Tools. You can expand the entry to see the available tools: `list_voices`, `get_voice`, `list_templates` and `get_template`.

<Tip>You can also place the configuration at `.cursor/mcp.json` inside a specific project directory to scope the connection to that project only. The global configuration at `~/.cursor/mcp.json` applies across all projects.</Tip>

## How to know it worked

Open a new Agent conversation in Cursor and send a prompt that requires brand voice context, for example: "Using Brivvy, get the brand voice rules and write a two-sentence product update." The Agent will call the appropriate Brivvy tool and return a response shaped by your brand voice rules rather than a generic output.


# Editor
Source: https://docs.brivvy.io/editor

Rich text editing to create modern communications.

<Frame>
  <img title="Editor Light" alt="Editor Cover Light" />

  <img title="Editor Light" alt="Editor Cover Light" />
</Frame>

The Brivvy editor supports rich text formatting within Comms. Markdown syntax and keyboard shortcuts provide quick formatting options for brand voice descriptions, template instructions and audience definitions.

## Overview

The editor converts Markdown syntax into rich text automatically. Users can paste Markdown directly or type using standard Markdown notation.

## Formatting

A formatting toolbar appears when text is highlighted. Slash commands accessed by typing <kbd>/</kbd> provide additional formatting options.

#### Text styling

* `**text**` or <kbd>Cmd/Ctrl</kbd> <kbd>B</kbd> for bold text
* `_text_` or <kbd>Cmd/Ctrl</kbd> <kbd>I</kbd> or <kbd>Cmd/Ctrl</kbd> <kbd>></kbd> for italicized text
* `~text~` or <kbd>Cmd/Ctrl</kbd> <kbd>Shift</kbd> <kbd>S</kbd> to strikethrough text
* <kbd>Cmd/Ctrl</kbd> <kbd>U</kbd> for underlined text
* <kbd>Cmd/Ctrl</kbd> <kbd>E</kbd> for inline code
* <kbd>#</kbd> then <kbd>Space</kbd> Heading 1
* <kbd>##</kbd> then <kbd>Space</kbd> Heading 2
* <kbd>###</kbd> then <kbd>Space</kbd> Heading 3

#### Lists

* <kbd>-</kbd> then <kbd>Space</kbd> or <kbd>Cmd/Ctrl</kbd> <kbd>Shift</kbd> <kbd>8</kbd> for a bulleted list
* `1.` or <kbd>Cmd/Ctrl</kbd> <kbd>Shift</kbd> <kbd>7</kbd> for a numbered list

#### Other formatting options

* <kbd>Cmd/Ctrl</kbd> <kbd>K</kbd> turn text into link (or directly paste issue or URLs for clickable links)
* <kbd>></kbd> then <kbd>Space</kbd> for blockquotes
* /code or <kbd>Cmd/Ctrl</kbd> <kbd>Alt</kbd> <kbd>C</kbd> for a code block
* <kbd>\_\_\_</kbd> then <kbd>Space</kbd> for a horizontal divider
* <kbd>/table</kbd> to create a table element

## Helpful commands

* <kbd>Cmd/Ctrl</kbd> <kbd>A</kbd> to select all content in a comm (to copy or delete)
* <kbd>Cmd/Ctrl</kbd> <kbd>Z</kbd> to undo typing
* <kbd>Cmd/Ctrl</kbd> <kbd>Shift</kbd> <kbd>Z</kbd> to redo typing
* <kbd>Shift</kbd> <kbd>Enter</kbd> to generate a line break
* <kbd>Enter</kbd> to break out of codeblock or blockquote formatting

## Copying content

Content can be copied in Markdown format for use in external tools. Select the desired content and use standard copy commands.

Markdown export preserves formatting including bold text, lists, code blocks and links.

## Editor behavior

The editor auto-saves content as changes are made. No manual save action is required.

Formatting persists across browser sessions and devices. All workspace members see consistent formatting when viewing shared content.

## Keyboard shortcuts reference

| Action        | Mac                                              | Windows/Linux                                     |
| ------------- | ------------------------------------------------ | ------------------------------------------------- |
| Bold          | <kbd>Cmd</kbd> + <kbd>B</kbd>                    | <kbd>Ctrl</kbd> + <kbd>B</kbd>                    |
| Italic        | <kbd>Cmd</kbd> + <kbd>I</kbd>                    | <kbd>Ctrl</kbd> + <kbd>I</kbd>                    |
| Strikethrough | <kbd>Cmd</kbd> + <kbd>Shift</kbd> + <kbd>S</kbd> | <kbd>Ctrl</kbd> + <kbd>Shift</kbd> + <kbd>S</kbd> |
| Inline code   | <kbd>Cmd</kbd> + <kbd>E</kbd>                    | <kbd>Ctrl</kbd> + <kbd>E</kbd>                    |
| Link          | <kbd>Cmd</kbd> + <kbd>K</kbd>                    | <kbd>Ctrl</kbd> + <kbd>K</kbd>                    |
| Bulleted list | <kbd>Cmd</kbd> + <kbd>Shift</kbd> + <kbd>8</kbd> | <kbd>Ctrl</kbd> + <kbd>Shift</kbd> + <kbd>8</kbd> |
| Numbered list | <kbd>Cmd</kbd> + <kbd>Shift</kbd> + <kbd>7</kbd> | <kbd>Ctrl</kbd> + <kbd>Shift</kbd> + <kbd>7</kbd> |
| Code block    | <kbd>Cmd</kbd> + <kbd>Alt</kbd> + <kbd>C</kbd>   | <kbd>Ctrl</kbd> + <kbd>Alt</kbd> + <kbd>C</kbd>   |
| Select all    | <kbd>Cmd</kbd> + <kbd>A</kbd>                    | <kbd>Ctrl</kbd> + <kbd>A</kbd>                    |
| Undo          | <kbd>Cmd</kbd> + <kbd>Z</kbd>                    | <kbd>Ctrl</kbd> + <kbd>Z</kbd>                    |
| Redo          | <kbd>Cmd</kbd> + <kbd>Shift</kbd> + <kbd>Z</kbd> | <kbd>Ctrl</kbd> + <kbd>Shift</kbd> + <kbd>Z</kbd> |


# Gemini
Source: https://docs.brivvy.io/gemini

Connect Brivvy to Gemini CLI so the model can access brand voice rules and generate on-brand content directly from the terminal.

Gemini CLI supports remote MCP servers via HTTP, which means you can connect Brivvy's MCP server with a single configuration entry. Once connected, Gemini can call Brivvy tools mid-conversation to retrieve brand voice guidelines and list templates without leaving the terminal.

## Before you start

* Gemini CLI installed (run `npm install -g @google/gemini-cli` if not already).
* A Brivvy account on a plan that includes MCP access.
* A text editor for modifying `settings.json`.

## Steps

1. Open a terminal and navigate to the Gemini CLI config directory. The path depends on your platform:

   | Platform      | Path                             |
   | ------------- | -------------------------------- |
   | macOS / Linux | `~/.gemini/settings.json`        |
   | Windows       | `%APPDATA%\gemini\settings.json` |

   Create the file if it does not exist.

2. Open `settings.json` in a text editor. The file uses standard JSON format. If the file is empty, start with an opening and closing curly brace `{}` as the base structure.

3. Paste the following configuration inside the top-level object. If a `mcpServers` block already exists, add `"brivvy"` as an additional entry rather than creating a second block.

   ```json theme={null}
   {
     "mcpServers": {
       "brivvy": {
         "httpUrl": "https://mcp.brivvy.io",
         "oauth": {
           "enabled": true
         }
       }
     }
   }
   ```

4. Save the file. Make sure the JSON is valid before closing the editor. A trailing comma after the last entry or mismatched brackets are common issues that will prevent Gemini CLI from starting.

5. Run `gemini` in the terminal. The CLI reads `settings.json` on startup and attempts to connect to each configured MCP server.

6. Inside the Gemini CLI session, run `/mcp auth brivvy` to start the OAuth flow. Your browser will open an authorization page where you can sign in to Brivvy and grant access.

7. After authentication completes, type `/mcp` and press <kbd>Enter</kbd>. The output lists all connected servers and their available tools. `brivvy` should appear with tools such as `get_voice` and `list_templates`.

8. Send a prompt that requires brand voice context, for example: "Using Brivvy, get the brand voice rules and write a two-sentence product update." Gemini CLI will call the appropriate Brivvy tool and incorporate the returned guidelines into the response.

## How to know it worked

The `/mcp` command shows `brivvy` with a connected status and a list of tools. A test prompt that references Brivvy returns a response shaped by the brand voice rules rather than a generic output.


# Glossary Overview
Source: https://docs.brivvy.io/glossary-overview

Manage your organization's terminology to keep product names, acronyms and key terms consistent across all content.

The glossary is where you define the terms your team should always use and the terms it should avoid. It works alongside your brand voice to enforce terminology at the point of content generation. When an AI tool connected through the MCP server creates content, it checks your glossary and applies the correct terms automatically.

You can think of the glossary as two lists. **Preferred terms** are the approved names, acronyms and phrases your organization has standardized on. **Avoid terms** are words and phrases your team should not use, each paired with a recommended alternative. Together, these lists eliminate inconsistency before it reaches your audience.

## How it works

When you add a term to the glossary, it becomes a constraint that AI tools enforce during content generation. Preferred terms are always used exactly as written. Avoid terms are automatically replaced with the alternatives you specify.

The glossary is workspace-wide. Every member of your workspace shares the same terminology, and any connected AI tool has access to it through the MCP server. Changes take effect the next time content is generated.

## Two term types

The glossary supports two types of entries, each with different behavior and configuration options.

**Preferred terms** define the correct way to reference a product, feature, concept or acronym. You can mark a preferred term as a proper noun, assign a short name for abbreviation handling and list related names that should not be used. For a full breakdown of preferred term fields and behavior, see [Preferred terms](/preferred-terms).

**Avoid terms** define language your team should not use. Each avoid term includes one or more recommended alternatives. When AI tools encounter a reason to use an avoided term, they substitute one of the alternatives instead. For details on how avoid terms work, see [Avoid terms](/avoid-terms).

## Adding a term

In your Brivvy workspace, navigate to the glossary. Select the type of term you want to add: preferred or avoid. Fill in the required fields and save. The term is immediately available to all connected AI tools.

## Best practices

* Keep entries concise. A glossary works best when each entry is specific and unambiguous.
* Review your glossary quarterly. Terminology changes as your product evolves, and outdated entries create confusion.
* Coordinate with your voice configuration. The glossary handles terminology while voice handles tone and style. Use both together for full coverage.
* Start with your most common pain points. If your team frequently misspells a product name or uses an outdated term, add those first.

## Related

* [Preferred terms](/preferred-terms) to learn how to configure terms your team should always use.
* [Avoid terms](/avoid-terms) to learn how to flag language your team should not use.
* [Voices](/voices) to configure the tone and rules that work alongside your glossary.
* [MCP Server](/mcp-server) to connect AI tools that enforce your glossary during content generation.


# Brivvy Docs
Source: https://docs.brivvy.io/index

Get an overview of Brivvy's features, integrations, and how to use them.

Brivvy is an intelligent communications platform that helps companies create on-brand content at scale. In these guides, you'll find everything you need to configure your brand voice, build content templates, manage audiences, and maintain consistency across all your communications.

Our docs cover setup instructions, best practices, feature references, and integration guides to help you get the most out of Brivvy, whether you're just getting started or building advanced workflows for your team.

<Columns>
  <Card icon="zap" href="/quickstart" title="Getting Started">
    Learn how to use the app and follow best practices for building comms at scale.
  </Card>

  <Card icon="audio-lines" href="/voices" title="Voices">
    Set up and configure your voice's tone, style, and writing guidelines.
  </Card>

  <Card icon="blocks" href="/create-template" title="Templates">
    Build reusable content structures for consistent communications.
  </Card>

  <Card icon="sparkles" href="/mcp-server" title="AI Tools">
    Access your brand voice and templates directly in popular AI clients.
  </Card>
</Columns>


# Invite members
Source: https://docs.brivvy.io/invite-members

Invite and manage members of your workspace.

You can invite members to your workspace via email or through a shareable link. Admins control who can send invitations and how new members join.

<iframe title="YouTube video player" />

<Callout icon="youtube">
  If you prefer video tutorials, check out our [YouTube](https://www.youtube.com/@brivvyHQ) page.
</Callout>

## Send an invitation

### Email invitation

1. Go to **Settings > Administration > Members**.
2. Click the **Invite** button.
3. Enter the invitee's email address. To add multiple invitees, separate each email with a comma.
4. Under **Invite as...**, select the role you want the invitee to have (paid plans only).
5. Select the team or teams you want the invitee to automatically join.
6. Click **Send invites**. New members receive an invite link by email along with steps to join the workspace.

<Note>
  If invitation emails are being filtered, add [notifications@brivvy.com](mailto:notifications@brivvy.com) to your allowlist as a trusted sender.
</Note>

### Share link

A share link lets anyone with the URL join your workspace without a direct invitation.

1. Go to **Settings > Administration > Members**.
2. Click **Invite**, then select the **Share link** tab.
3. Copy the link and share it internally with your organization.

You can disable sharing at any time, which restricts workspace access to email invitations only. The link can also be regenerated at any time, replacing the previous one.

## Invite and assign

You can assign invited users to comms or mark them as project leads before they accept their invitation.

On any comm, open the assignee selection menu and choose **Invite and assign...**. After inviting them, search for them in the assignee menu to continue allocating work.

Or, when sending an invite from the workspace settings members page, click the invited user and create new comms from their user page. These will automatically be assigned to them.

## Email options by plan

<Tabs>
  <Tab title="Free plan">
    All members of a workspace on the free plan are considered admins. Anyone can send invitations to new members.
  </Tab>

  <Tab title="Paid plans">
    On paid plans, members can be assigned the admin or member role. Only admins can invite new members by default.
  </Tab>
</Tabs>


# MCP Server
Source: https://docs.brivvy.io/mcp-server

Learn how to connect Brivvy's MCP Server with Claude's web-based client, desktop client, and Claude Code.

The Brivvy Model Context Protocol (MCP) server lets AI assistants like Claude access your brand voice, content templates, audience definitions, and glossary terms directly. This means you can generate on-brand content right inside your AI tools.

When connected, any MCP-compatible AI application can use Brivvy's brand intelligence to create content that matches your voice and style guidelines. The server provides secure, authenticated access to your workspace through a standardized protocol.

Brivvy's MCP server is hosted and maintained by us, following the authenticated remote [MCP specification](https://modelcontextprotocol.io/specification/2025-03-26). The available tools let you retrieve brand voice settings, access content templates, query audience definitions, and look up glossary terms. This ensures every piece of AI-generated content aligns with your brand standards.

## General

The Brivvy MCP server supports the Streamable HTTP transport. Authentication uses OAuth 2.1 with dynamic client registration at:

```text theme={null}
https://mcp.brivvy.io
```

## Available tools

The Brivvy MCP server provides the following [tools](https://modelcontextprotocol.io/specification/2025-11-25/server/tools):

| Resource  | Tool                                     | Description                                                               |
| --------- | ---------------------------------------- | ------------------------------------------------------------------------- |
| Voices    | <Badge>list\_voices</Badge>              | Returns all brand voices available in your workspace.                     |
| Voices    | <Badge>get\_voice</Badge>                | Retrieves the full tone and style rules for a specific brand voice.       |
| Templates | <Badge>list\_templates</Badge>           | Returns all content templates saved in your workspace.                    |
| Templates | <Badge>get\_template</Badge>             | Retrieves the generation prompt and instructions for a specific template. |
| Templates | <Badge>list\_discover\_templates</Badge> | Browses publicly available templates from the Brivvy community.           |

## Security and authentication

The Brivvy MCP Server uses OAuth authentication to connect with the Brivvy platform. Authentication tokens are stored locally on the device running the AI platform.

No brand data is transmitted to AI platforms directly. The AI platform requests data from the MCP Server as needed during content generation. All requests require valid authentication.

## FAQs

<AccordionGroup>
  <Accordion title="I'm seeing an internal server error when connecting. What should I do?">
    Clear your saved authentication data by running this in Terminal: `rm -rf ~/.mcp-auth` and try reconnecting. You may also need to update to a newer version of Node.js if prompted.
  </Accordion>

  <Accordion title="I'm using WSL on Windows and encountering errors.">
    Try this alternative configuration:

    ```text theme={null}
    {
      "mcpServers": {
        "brivvy": {
          "command": "wsl",
          "args": ["npx", "-y", "mcp-remote", "https://mcp.brivvy.io"]
        }
      }
    }
    ```
  </Accordion>

  <Accordion title="Which workspace will the MCP server access?">
    The server accesses the workspace tied to your authentication credentials. If you have multiple workspaces, specify which one to use during OAuth or in your Brivvy account settings.
  </Accordion>
</AccordionGroup>


# Members and roles
Source: https://docs.brivvy.io/members-roles

Manage workspace members and understand the available role types.

Brivvy provides role types to help you control access and permissions across your workspace. Each role gives team members the right level of access, from full administrative control to standard collaboration.

<iframe title="YouTube video player" />

<Callout icon="youtube">
  If you prefer video tutorials, check out our [YouTube](https://www.youtube.com/@brivvyHQ) page.
</Callout>

## Overview

Admins can manage workspace members from **Settings > Administration > Members**. This page lists all active and suspended members and allows filtering by role or status, including pending invites, suspended users or members who have left the workspace.

## Managing user roles

### Changing a member's role

To update a user's role:

1. Go to **Settings > Administration > Members**.
2. Hover over a member's row.
3. Click the **overflow menu (⋯)**.
4. Select **Change role...**.

### Suspend a member

Admins can suspend a member from the workspace:

1. Go to **Settings > Administration > Members**.
2. Hover over a member's row.
3. Click the **overflow menu (⋯)**.
4. Select **Suspend user...**.

Suspended users lose all access immediately and are removed from your next billing cycle. They remain visible in the members list for historical purposes, for example when viewing comms they created or were assigned to.

## Role types

<Note>
  On the free plan, all workspace members are automatically assigned the admin role. Role-based access control becomes available when you upgrade to a paid plan.
</Note>

### Admin

Admins have elevated permissions to manage routine workspace operations. This role is well-suited for managers, team leads and operations-focused members.

On paid plans, the user who upgrades the workspace is granted the admin role by default.

### Member

Members can collaborate across the workspace and use all standard features. They cannot access workspace-level administration pages.


# Preferred Terms
Source: https://docs.brivvy.io/preferred-terms

Define the correct names, acronyms and phrases your team should always use in content.

Preferred terms are the standardized names your organization has agreed on. When you add a preferred term to the [glossary](/glossary-overview), AI tools connected through the MCP server will use that term exactly as written during content generation. This is how you keep product names, feature names and key concepts consistent without relying on manual review.

## When to use preferred terms

Add a preferred term when your organization has an official name for something and you need it used consistently. Common examples include product names, feature names, industry-specific terminology and acronyms that appear frequently in your content.

## Fields

Each preferred term supports the following fields.

**Name.** The full, official name of the term. This is the exact string that AI tools will use in generated content. Capitalization matters, so enter the term exactly as it should appear.

**Proper noun.** A toggle that tells AI tools whether to always capitalize the term. Enable this for brand names, product names and any other term that should never appear in lowercase. When a term is marked as a proper noun, it will always be capitalized regardless of where it appears in a sentence.

**Short name.** An optional abbreviation or acronym for the term. When you provide a short name, AI tools follow a specific pattern: the first mention in any piece of content uses the full name followed by the short name in parentheses, and every subsequent mention uses only the short name. For example, if the name is "application programming interface" and the short name is "API," the first reference reads "application programming interface (API)" and all later references read "API."

**Never abbreviate.** A toggle that overrides short name behavior. When enabled, AI tools always use the full name and never substitute the short name, even after the first mention. Use this when the full term is important for clarity or when your audience may not recognize the abbreviation.

**Related names.** A list of incorrect or outdated variants that people sometimes use instead of the preferred term. AI tools treat related names as blocked alternatives. If content would naturally include one of these variants, the preferred name is used instead. For example, if the preferred term is "Brivvy" and a related name is "brivvy," the lowercase version will never appear in output.

## Examples

Here are a few examples to illustrate how preferred terms work in practice.

**Product name.** Name: "Brivvy." Proper noun: enabled. No short name. This ensures the product name is always capitalized and never appears as "brivvy" or any other variant.

**Technical acronym.** Name: "Model Context Protocol." Proper noun: enabled. Short name: "MCP." On first mention, content reads "Model Context Protocol (MCP)." Every following mention reads "MCP."

**Common abbreviation.** Name: "frequently asked questions." Short name: "FAQs." Proper noun: disabled. First mention reads "frequently asked questions (FAQs)." Subsequent mentions read "FAQs."

## Best practices

* Enter names with the exact capitalization you want in your content. AI tools reproduce the name character for character.
* Use the proper noun toggle for any term that should always be capitalized. Do not rely on AI tools to infer capitalization.
* Add related names for every known variant or misspelling. The more variants you capture, the fewer inconsistencies reach your audience.
* Set a short name only when the abbreviation is well known to your audience. If the acronym might cause confusion, enable never abbreviate instead.
* Keep preferred terms focused on terminology, not style. Use your [voice configuration](/voices) for tone and formatting rules.

## Related

* [Glossary](/glossary-overview) for an overview of how the glossary works.
* [Avoid terms](/avoid-terms) to define words and phrases your team should not use.
* [Voices](/voices) to configure tone and rules that work alongside your terminology.


# Preview voice
Source: https://docs.brivvy.io/preview-brand-voice

Test how brand voice settings shape AI output before applying them to your communications.

<Frame>
  <img title="Preview Brand Voice Light" alt="Preview Brand Voice Cover Dark" />

  <img title="Preview Brand Voice Light" alt="Preview Brand Voice Cover Light" />
</Frame>

Preview brand voice lets you see the effect of your configuration in real time, before any content is created. It is a practical way to fine-tune settings and understand how small adjustments change the way AI writes on behalf of your organization.

## How to preview a brand voice

1. In your Brivvy workspace, navigate to the **Brand Voice** tab.
2. Select the brand voice you want to preview, or create a new one if needed.
3. Select **Preview** to open the preview panel alongside the configuration.
4. Make changes to Tone dimensions or Rules. The preview updates in real time to reflect each adjustment.
5. Once satisfied with the output, select **Publish** to apply the configuration across the workspace.

<Note>
  To regenerate the preview, you must make a change to Tone or Rules. The preview does not refresh automatically on open.
</Note>

## What preview shows

Preview reflects the two core components of a brand voice: Tone and Rules.

Tone controls the emotional character of the writing. Each dimension is adjustable on a scale of 1 to 10:

| Dimension       | Low (1)                    | High (10)             |
| --------------- | -------------------------- | --------------------- |
| Warmth          | Detached, clinical         | Friendly, personable  |
| Confidence      | Tentative, hedging         | Assertive, definitive |
| Formality       | Casual, conversational     | Formal, structured    |
| Playfulness     | Serious, measured          | Light, witty          |
| Technical depth | Accessible, plain language | Precise, expert-level |

Rules define the specific linguistic and stylistic constraints applied during generation. These span:

| Category                 | Examples                                                    |
| ------------------------ | ----------------------------------------------------------- |
| Linguistic behavior      | Voice, perspective, use of contractions and casual phrasing |
| Punctuation              | Dashes, commas, semicolons, exclamation marks               |
| Formatting and structure | Heading case, sentence length, paragraph length             |
| Spelling variants        | Preferred spelling and regional conventions                 |
| Dates, times and numbers | Formatting conventions for each                             |

## Preview theme

The preview theme lets you set an industry context for the generated output. Adding a theme helps Brivvy produce more realistic sample content, making it easier to evaluate how your brand voice settings will feel in practice.

To set a theme, select a short industry or use case in the **Theme** field before opening the preview. Examples include "fintech", "e-commerce", "productivity tool" or "B2B SaaS". The preview will incorporate that context into the generated content alongside your workspace name.

Themes are optional. Without one, Brivvy generates a generic sample based on your configuration alone.


# Preview template
Source: https://docs.brivvy.io/preview-template

See how your template prompt shapes content in real time before publishing.

Preview template lets you test your template prompt as you build it. Provide a description and a prompt, and Brivvy generates a live content preview using your workspace name and any theme context you add. As you refine the prompt, the preview updates automatically.

Templates define structure, not voice. The prompt shapes how content is organized and formatted, not how it sounds. Brand voice handles tone and style separately.

## How it works

The preview uses three inputs to generate a sample output:

* **Description,** a brief summary of what the template is for
* **Prompt,** the structural instructions that define sections, formatting and content guidelines
* **Theme,** optional context that helps the preview reflect a realistic use case, such as "productivity tool", "fintech" or "e-commerce"

As you edit the prompt, the preview refreshes to show how your structural instructions will shape generated content.

## Steps

1. Navigate to **Templates** and open an existing template or create a new one.
2. Enter a short description of the template's purpose. This helps orient the preview and gives context to the generated output.
3. In the **Prompt** field, write the structural instructions for your template. Define sections, specify formatting and set any content requirements.
4. Review the content sample Brivvy generates. The preview uses your workspace name and theme to produce a realistic output shaped by your prompt.
5. Adjust the prompt and watch the preview update in real time. Continue refining until the structure matches your intent.

## Preview scope

The preview shows how your prompt structures content. It does not apply your Brand Voice. Tone, punctuation rules and linguistic behavior are applied separately when the template is used to generate a comm.

Use the preview to validate structure, such as section order, heading levels and content guidelines, rather than to evaluate voice or style.


# Preview template
Source: https://docs.brivvy.io/preview-templates





# Profile
Source: https://docs.brivvy.io/profile

Learn how to navigate and manage your Brivvy profile.

## Overview

Your profile found in [Settings > Profile settings > Profile](https://app.brivvy.io/settings/profile) is where you personalize how other team members will see you in your Brivvy workspace.

## Set your profile picture

The default avatar will be the first and last initials of your account. To add a new photo, hover over the avatar until the edit pencil icon shows up, and click on it. A pop-up folder will show up for you to select the photo you want to update.

Once you've updated your avatar, you cannot remove it—only replace it with another photo.

## Name and display name

Members can set whether they see their team members' full names or usernames in their **Settings > Profile settings > Profile**. You can adjust these to your preference of how you want others to refer you to.

## Email address

Your email address is your unique ID (User Account) for all workspaces you have created a User for. For example:

| Action                                                                                   | Result                                                                                                                                                                                         |
| ---------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| You joined Workspace A using [jane.doe@email.com](mailto:jane.doe@email.com)             | A User Account was created under [jane.doe@email.com](mailto:jane.doe@email.com) and a User was created for Workspace A.                                                                       |
| You joined Workspace B.                                                                  | A User was created for Workspace B under your User Account [jane.doe@email.com](mailto:jane.doe@email.com)                                                                                     |
| You updated your email address in Workspace B to [jdoe@email.com](mailto:jdoe@email.com) | Your email address for your User Account, Workspace A, and Workspace B was updated to [jdoe@email.com](mailto:jdoe@email.com). [jane.doe@email.com](mailto:jane.doe@email.com) does not exist. |

### Change your email address

To change your email address:

1. Go to **Settings > Account > Profile**.
2. Hover over your email address and click the pencil icon.
3. A pop-up box will appear for you to enter the new email address you want to switch to.
4. A confirmation email will be sent to both old and new email addresses. You need to click *both* in order for the new email address to be confirmed.

> **Note:** Changing an email address will change the email address for **all** workspaces using the old email address. If you want to change the email address for just one workspace, we suggest inviting that email as a new user and leaving the workspace from the old email user.

If your organization is changing domains and you'd like to change all emails in your workspace, please reach out to us at [support@brivvy.com](mailto:support@brivvy.com) from the existing domain and let us know the domain you would like to change to.

## Leave your workspace

To leave the workspace, click the **Leave workspace** option in **Settings > Profile**. Once removed, you no longer have access to the workspace. An admin will have to unsuspend your account in order for you to log back in.

<Note>
  You cannot leave your workspace if you are the last admin. You must first assign another user as an admin before leaving.
</Note>


# Quickstart
Source: https://docs.brivvy.io/quickstart

Get started with Brivvy's intelligent communications platform.

Brivvy helps teams create on-brand content at scale. This guide covers the essential setup steps to get your workspace ready and your team generating consistent, on-brand content quickly.

## Core features

* [Voices](/voices) — Configure tone and writing rules that guide AI-generated content
* [Templates](/create-template) — Build reusable structures for consistent comm output
* [MCP Server](/mcp-server) — Connect AI tools like Claude, Cursor and ChatGPT to your brand assets

## Setup

<Tabs>
  <Tab title="For admins">
    Admins should complete these steps to prepare the workspace for team content generation. The order matters — brand voice and templates should be in place before the team starts creating comms.

    1. Navigate to [Voices](/voices) and set the five tone sliders to match the desired communication style. Then add writing rules that define formatting preferences, perspective and style requirements.
    2. Build templates for the most common content types the team produces. Start with three to five high-priority templates such as blog posts, feature announcements or social media content. See [create a template](/create-template) for guidance.
    3. Set up [MCP Server](/mcp-server) connections in the AI platforms the team uses. Follow the platform-specific guides for [Claude](/claude), [Cursor](/cursor) or [ChatGPT](/chatgpt) to enable workspace access.
    4. [Invite team members](/invite-members) and assign roles so the workspace is ready for collaboration.
  </Tab>

  <Tab title="For members">
    Once an admin has configured the workspace, members can get up to speed quickly. The goal is to understand the brand voice and templates before creating your first comm.

    1. Open the [Voices](/voices) page to understand tone configuration and writing rules. This gives you important context for how content should sound and be formatted.
    2. Browse [Templates](/create-template) to familiarize yourself with the content structures available for your team.
    3. Connect your account to the [MCP Server](/mcp-server) in the AI platforms you use. Follow the platform-specific guides for [Claude](/claude), [Cursor](/cursor) or [ChatGPT](/chatgpt).
  </Tab>
</Tabs>


# Share comm
Source: https://docs.brivvy.io/share-comm

Generate a public link to share any comm outside your workspace, with full control over access.

<Frame>
  <img title="Share Comm Light" alt="Share Comm Cover Dark" />

  <img title="Share Comm Light" alt="Share Comm Cover Light " />
</Frame>

Comms can be shared externally as a read-only public page. This is useful for sending a polished communication to people outside an organization (investors, clients, agencies, partners or any third party) without giving them access to the workspace. Comms can be shared internally with workspace members or externally with anyone outside the organization. External sharing generates a public link to a read-only view of the comm — no Brivvy account required to view it.

## Share a comm externally

Any workspace member can share a comm externally.

<Steps>
  <Step title="Open the comm">
    Navigate to the comm to share.
  </Step>

  <Step title="Open the share menu">
    Click **Share** in the top right of the comm.
  </Step>

  <Step title="Enable external access">
    Under **External access**, open the dropdown and select **Anyone with link**. Brivvy automatically generates a public link.
  </Step>

  <Step title="Copy and share the link">
    Click **Copy link** and send it to recipients.
  </Step>
</Steps>

### What recipients see

Recipients who open a public comm link see a clean, read-only page with the full comm content. They cannot edit the comm, leave comments or access any other part of the workspace. An **Open app** button is visible to anyone who is already a workspace member.

<Frame>
  <img title="Disable External Access Light" alt="Disable External Access Cover Dark" />

  <img title="Disable External Access Light" alt="Disable External Access Cover Light" />
</Frame>

## Manage access

### Disable external access

To revoke access, open the share menu and change the **External access** dropdown back to **Not shared**. The link is disabled immediately and anyone who previously had it will no longer be able to view the comm.

### Regenerate the link

To invalidate an existing link without disabling access entirely, click the regenerate icon next to the link field. Brivvy generates a new link and the previous one stops working immediately.

<Frame>
  <img title="Regenerate Link Light" alt="Regenerate Link Cover Dark" />

  <img title="Regenerate Link Light" alt="Regenerate Link Cover Light" />
</Frame>

<Note>
  Regenerating a link cannot be undone. Anyone with the old link loses access as soon as the new link is generated.
</Note>

## Plan availability

External sharing is available on all Brivvy plans.


# Create a voice from scratch
Source: https://docs.brivvy.io/start-from-scratch

Define your brand voice manually by setting tone dimensions and writing rules.

Creating a voice from scratch gives you full control over how your brand sounds. You set every parameter yourself, from the emotional character of your writing down to how you format dates and numbers.

This is the best approach when you are building a voice from the ground up or when you need precise control over specific settings.

## Before you start

* Any member of your workspace can create a voice. There are no role restrictions.
* It helps to have a clear picture of your brand's communication style before you begin. Think about who your audience is, what tone fits your product and which writing conventions your team already follows.

## How to create a voice from scratch

1. In your Brivvy workspace, navigate to the **Voices** tab.
2. Select **New brand voice**. If you already have voices in your workspace, this appears as a dropdown. Choose **Start from scratch**.
3. Give your voice a clear name that reflects its purpose, for example: Product, Marketing, Social or Founder.
4. Set each [tone dimension](#tone) using the 1–10 sliders.
5. Work through each [rules category](#rules) and configure the settings to match your organization's standards.
6. Use [Preview](/preview-brand-voice) to see how your configuration shapes AI output before going live.
7. Select **Publish** to make the voice available across your workspace.

## Tone

Tone defines the emotional character of your writing. Five dimensions are adjustable on a scale of 1 to 10. Each dimension is independent, so you can combine them in whatever way fits your brand.

| Dimension           | Low (1)                    | High (10)             |
| ------------------- | -------------------------- | --------------------- |
| **Warmth**          | Detached, clinical         | Friendly, personable  |
| **Confidence**      | Tentative, hedging         | Assertive, definitive |
| **Formality**       | Casual, conversational     | Formal, structured    |
| **Playfulness**     | Serious, measured          | Light, witty          |
| **Technical depth** | Accessible, plain language | Precise, expert-level |

Start by setting tone before you move to rules. Tone establishes the overall character of your voice. Rules then handle the specific constraints that keep output consistent.

## Rules

Rules define the linguistic and stylistic constraints applied during content generation. They are organized into eight groups, each covering a different aspect of your writing standards.

Brivvy shows examples and explanations for every option inside the product, so you can see exactly what each setting does as you configure it.

### Linguistic behavior

Controls the foundational writing choices that shape how your content reads.

* **Pronoun perspective.** Whether your writing uses first person ("we"), second person ("you"), third person or a neutral style.
* **Sentence voice.** Active, passive or a preference for one over the other.
* **Slang level.** How much casual or colloquial language is allowed.
* **Emoji frequency.** Whether emojis are used and how often.
* **Gender-neutral wording.** Whether to enforce inclusive, gender-neutral language.
* **Language sensitivity.** The level of care applied to identity-related and inclusive wording.

### Punctuation

Controls how specific punctuation marks are used across your content. This is the most detailed rules group, covering items like:

* **Oxford comma.** Whether to include the final comma in a list of three or more items.
* **Colon usage.** Where colons are allowed, such as lists only, explanations or both.
* **Parenthetical style.** How asides are set off: dashes, parentheses, commas or avoided entirely.
* **Trailing break style.** How to handle emphasis at the end of a sentence.
* **Range and compound style.** Whether to use hyphens, en dashes or prose for ranges.
* **Ellipsis style.** Whether ellipses are used and which character format to follow.
* **Contractions.** How frequently contractions appear in your writing.
* **Ampersand usage.** Whether "&" is allowed or always spelled out as "and."
* **Semicolons.** Whether semicolons are used and how often.
* **Exclamation marks.** Whether exclamation marks are used and how often.
* **Plus-sign usage.** Whether "+" can substitute for "and."
* **Quotation marks.** Double or single quotes, and where to place punctuation relative to closing quotes.

### Formatting and structure

Controls the visual shape and readability of your content.

* **Heading case.** Sentence case, title case, all caps or all lowercase.
* **Heading length.** Concise, standard or expanded headings.
* **Sentence length.** Concise (under 15 words), standard (15–25 words) or expanded (25–40 words).
* **Paragraph density.** How many sentences per paragraph: concise (1–2), standard (2–4) or expanded (4–6).
* **List display style.** Inline lists, bulleted lists or a hybrid approach.
* **List item punctuation.** How list items are terminated: no punctuation, periods, semicolons or matched to item type.
* **Link formatting.** Markdown links or plain text URLs.

### Spelling variants

* **Language locale.** American English or British English.
* **Simplified spelling.** Whether to prefer shorter variants when multiple spellings exist, such as "acknowledgment" over "acknowledgement."

### Dates

* **Date format.** Numeric (MM/DD/YYYY, DD/MM/YYYY, YYYY-MM-DD) or written (Month DD, YYYY or DD Month, YYYY).
* **Month style.** Full month names or abbreviated.
* **Leading zeros.** Whether single-digit months and days include a leading zero in numeric formats.

### Time

* **Time format.** 12-hour or 24-hour clock.
* **AM/PM style.** Lowercase with periods (a.m./p.m.) or uppercase without (AM/PM).
* **Timezone display.** Whether to include the timezone when stating times.

### Numbers

* **Thousands separator.** Commas, periods, spaces or none.
* **Numbers under 10.** Spelled out ("seven") or digits ("7").
* **Ordinal style.** Numeric (1st, 2nd) or spelled out (first, second).
* **Large-number abbreviations.** Full numbers, K/M/B shorthand or abbreviated in headings only.

### Units

* **Measurement system.** Metric or imperial.
* **Unit style.** Abbreviated symbols (km, kg) or full words (kilometers, kilograms).
* **Temperature.** Celsius or Fahrenheit.

## Preview

Before publishing, use the preview panel to see how your tone and rules shape AI output in real time. The preview updates as you make changes, so you can fine-tune settings and see the effect immediately.

For a detailed walkthrough, see [Preview brand voice](/preview-brand-voice).

## Publishing

Select **Publish** to make the voice available across your workspace. Once published, the voice can be used in any AI client connected through the [MCP server](/mcp-server) and is accessible to all workspace members.

You can edit and republish a voice at any time. Changes take effect the next time the voice is applied to content generation.

## Tips

* **Start with tone, then refine rules.** Tone sets the overall character. Rules handle edge cases and consistency. Getting tone right first makes rule decisions more intuitive.
* **Do not configure everything at once.** Focus on the settings that matter most to your brand. You can always come back and refine.
* **Use preview as you go.** Checking preview after each group of changes helps you understand the impact of individual settings.
* **Name your voice clearly.** Use a name that reflects purpose or audience, such as "Product," "Marketing" or "Support." This helps team members select the right one without guessing.

## What's next

* [Preview brand voice](/preview-brand-voice) to test your configuration in detail.
* [Add existing voice](/add-existing-voice) if you prefer to extract a voice from examples instead.
* [Connect an AI tool](/claude) to start generating content with your new voice.


# Voices Overview
Source: https://docs.brivvy.io/voices

Configure your organization's tone and writing rules to keep every communication on-brand.

Voices is where your team defines how it sounds. Set it up once in Brivvy and it becomes the standard that shapes every piece of content you create, whether you are working inside Brivvy or using an AI client through the [MCP server](/mcp-server).

A voice is made up of two components: **Tone** and **Rules**. Together, they give Brivvy a precise, consistent definition of how your organization communicates.

<iframe title="YouTube video player" />

<Callout icon="youtube">
  If you prefer video tutorials, check out our [YouTube](https://www.youtube.com/@brivvyHQ) page.
</Callout>

## Two components

**Tone** defines the emotional character of your writing. Five dimensions are adjustable on a scale of 1 to 10: Warmth, Confidence, Formality, Playfulness and Technical depth.

**Rules** define the specific linguistic and stylistic constraints applied during content generation. These cover linguistic behavior, punctuation, formatting and structure, spelling variants, dates, times, numbers and units.

## Create a voice

There are two ways to create a voice in Brivvy. Choose the approach that fits where you are with your brand.

* [**Start from scratch**](/start-from-scratch)**.** Define tone and rules manually. Best when you are building a voice from the ground up or want full control over every setting.
* [**Add existing voice**](/add-existing-voice)**.** Upload examples of your writing and let Brivvy extract a voice profile automatically. Best when your brand already has published content or style guides.

## Multiple voices

Brivvy supports multiple voices in a single workspace. This is useful for organizations that communicate differently across product lines, audiences or functions.

Common examples include a product voice (precise, technical), a marketing voice (warm, benefit-focused) and a social voice (concise, conversational).

You can set one voice as your **default**, which applies automatically when no specific voice is selected.

## Apply a voice

When working in an AI client that supports the Brivvy MCP server, specify which voice directly in your prompt. For example:

```text theme={null}
Apply the Marketing voice from Brivvy.
```

If you do not specify a voice, the <Badge>Primary</Badge> voice is used. If you name a specific voice, Brivvy applies that one exactly.

## Permissions

Any member of your workspace can create, edit, duplicate or delete voices. All published voices are available to everyone in the workspace.

<Note>
  Granular permissions are coming soon. This will let you control who can edit a voice and who can apply it when creating content.
</Note>

## Best practices

* **Name your voices clearly.** Use names that reflect purpose or audience so team members can select the right one without guessing.
* **Start with tone, then refine rules.** Tone sets the overall character. Rules handle edge cases and consistency. Getting tone right first makes rule decisions more intuitive.
* **Fewer, well-defined voices are better.** Too many similar voices create inconsistency rather than preventing it. Consolidate where you can.
* **Review periodically.** Brands evolve, and so does their voice. Revisit your configuration quarterly to make sure it still reflects how your organization communicates.


# Windsurf
Source: https://docs.brivvy.io/windsurf

Connect the Brivvy MCP Server with Windsurf's Cascade agent to access brand voice and templates while you code.

Windsurf's Cascade agent supports remote Model Context Protocol (MCP) servers over the Streamable HTTP transport. You can add Brivvy by dropping a single entry into your MCP configuration file. Once connected, Cascade can call Brivvy tools to retrieve brand voice rules, glossary terms and template instructions without leaving the editor. This is useful for generating on-brand README files, API documentation, code comments and user-facing copy right where you work.

## Before you start

* Windsurf installed and signed in to your account.
* A Brivvy account on a plan that includes MCP access.
* Node.js 18 or later installed if you plan to edit the config file directly.

## Steps

1. **Open the Cascade panel in Windsurf.** Click the Cascade icon in the top-right corner, or press Cmd+Option+B on macOS or Ctrl+Alt+B on Windows to toggle the sidebar.

2. **Open the MCP configuration.** Click the hammer icon in the Cascade panel, then click **Configure**. This opens the Manage plugins view. Click **View raw config** to open `mcp_config.json` in the editor. On macOS and Linux, this file lives at `~/.codeium/windsurf/mcp_config.json`. On Windows, it lives at `%USERPROFILE%\.codeium\windsurf\mcp_config.json`.

3. **Add the Brivvy entry.** Paste the following configuration into the file. If an `mcpServers` block already exists, add `brivvy` as an additional entry rather than creating a second block:

```json theme={null}
   {
     "mcpServers": {
       "brivvy": {
         "serverUrl": "https://mcp.brivvy.io"
       }
     }
   }
```

4. **Save the file.** Press Cmd+S on macOS or Ctrl+S on Windows to write your changes.

5. **Refresh the MCP connection.** Return to the Manage plugins view and click **Refresh**. Windsurf will load the new configuration and prompt you to authenticate.

6. **Authenticate with Brivvy.** Your browser will open an authorization page. Sign in to Brivvy and click **Authorize** to grant Cascade access. After you approve, Windsurf stores the OAuth credentials locally.

7. **Confirm the connection.** Back in Windsurf, check that Brivvy shows a connected status in the Manage plugins view. Expand the entry to see the available tools: `list_voices`, `get_voice`, `list_templates` and `get_template`.

## How to know it worked

Open a new Cascade conversation and send a prompt that requires brand voice context, for example: "Using Brivvy, get the brand voice rules and write a two-sentence product update." Cascade will call the appropriate Brivvy tool and return a response shaped by your brand voice rules rather than a generic output.


# Workspaces
Source: https://docs.brivvy.io/workspaces

Create and manage workspaces in Brivvy.

A workspace is the home for all comms, brand voices and templates for your organization.

## Overview

Your workspace brings together everything your team needs to communicate consistently, from brand voice configuration to comm creation.

Brivvy recommends staying within a single workspace, as this is the conceptual model used when designing the product.

## Workspace settings

To navigate to workspace settings, click your workspace name in the upper left and select **Settings**.

As a member, you will see settings related to your account, including notification preferences and personal details.

As an admin, you will also see an **Administration** section that is not viewable by members.

From **Workspace > Administration**, you can:

* Update the workspace name and URL
* Invite new members and manage existing members
* Manage billing and plan information
* Delete the workspace

## Delete workspace

Deleting a workspace removes all workspace data, including comms, brand voices and templates. This action is not reversible.

To delete your workspace, go to **Settings > Administration > General**.

<Warning>
  Workspace deletion is permanent. All data will be lost and cannot be recovered.
</Warning>

## Multiple workspaces

You can create multiple workspaces under a single account, tied to a specific email address. Each workspace has its own member list, brand voices and billing.

<Note>
  Adding a workspace from a different email account is not currently supported. This feature is coming soon.
</Note>