Skip to main content

FAQs

Getting started​

What is Revelry Ai?​

Revelry Ai gives your team an AI assistant that works from your own documents and the systems you already run. A project manager at a general contractor can ask which RFIs are still open on a job. It pulls them from Procore, checks the relevant specs in Knowledge, and drafts a status report the team can edit, share, and export.

You talk to Revelry Ai in Chat. Instructions your team repeats become Skills, which carry their own inputs, model, and tools, so anyone can run them the same way. Agents are assistants set up for one job, with their own instructions and Skills. Workflows chain Skills into multi-step processes. Everything it produces is saved as an Artifact with its full version history.

How do I get started?​

Enter your email on the sign-in page at app.revelry.ai. A new email gets an account, and you finish by opening the link sent to your email. For help setting up, the same page links to a short guided chat that asks about your business, uses your answers to name your organization and set up its first project, and saves the chat in that project. The Getting Started guide covers what to do next.

How do I sign in?​

Enter your email on the sign-in page, and you get a one-time sign-in link by email. Depending on how your deployment is set up, you can also sign in with Google or GitHub. Enterprise customers can use single sign-on (SSO) as well. See Authentication.

Can I belong to multiple organizations?​

Yes. Switch between them with the organization switcher at the bottom of the left navigation, below your name. Each organization has its own projects, Knowledge, integrations, and settings.

Features​

What can Revelry Ai work with?​

Your Knowledge holds uploaded documents and content synced from data sources such as GitHub, Notion, Google Sheets, and website crawls. Integrations connect live services such as Procore, Google Workspace, Microsoft 365, HubSpot, Pipedrive, and Slack. Chat can also search the web. The Chat Tools menu controls which of these are on for each conversation.

What AI models does Revelry Ai support?​

It works with seven model providers:

  • OpenAI
  • Anthropic (Claude), including Opus 5.5 and Sonnet 5.5, with Adaptive Thinking on newer Claude models
  • Google (Gemini)
  • Groq
  • Mistral
  • SpaceXAI (Grok)
  • Azure (OpenAI)

Your organization admin chooses which models are available and can set thinking or reasoning effort per model. Newer Claude models use Adaptive Thinking, and the other providers use their own reasoning settings. An individual Skill can pin a specific model. Wherever you choose a model, the options are grouped by provider and listed alphabetically.

What CRM integrations are available?​

HubSpot and Pipedrive. Each has to be turned on for your organization, so contact your Revelry Ai team to enable them. An organization admin then connects either one from the Shared tab of the Integrations page, and each person can also connect their own HubSpot account from the Personal tab. Both work in chat. HubSpot queries your CRM live during the conversation and does not copy anything into Knowledge. Pipedrive can read and update deals, contacts, organizations, activities, and notes.

Can I use Revelry Ai from my own applications?​

Yes. The API can create artifacts, run Workflows, hold chat conversations with streamed responses, and manage projects and other resources. Authenticate with an organization or personal API key.

Your work and your data​

Who can see my chats and artifacts?​

Chat History lists only your own conversations. When Revelry may share your data, including disclosures required by law, is set out in the Privacy Policy. Artifacts follow your organization's Groups: members see an artifact when one of their Groups has access to its project, and to its category if it has one. Share gives you a link to the artifact.

What happens if Revelry Ai gets something wrong?​

It asks before it changes anything in a connected service. Before it creates, updates, or deletes a record through an integration, it shows an approval prompt where you can choose Allow for this chat to let that tool run without asking for the rest of the chat, Allow Once, or Deny. Read-only lookups run without asking. Inside Revelry Ai, every change to an artifact records a version, so you can restore an earlier one, and Skills keep a version history too.

Can I export my data?​

You can copy any artifact, download it as a PDF, or send it to Google Drive once you've connected your Google account. You can also send it to GitHub as an issue once your organization adds a GitHub repository as a data source. For a full export of your account data, including chat history, email support@revelry.ai. The Privacy Policy covers what the export includes.

How do credits work?​

Your organization spends credits as it uses the platform. Model usage is charged by the tokens each request uses, and some actions, such as web searches, document processing, and Workflow runs, have their own cost. Organization Admins see the balance at the bottom of the left navigation. Other members see a notice there only when the balance runs out.

Do I need to enter billing information?​

Not on the Start plan. An organization admin can upgrade to Launch, Accelerate, or Scale without contacting us: click your credit balance in the left navigation to open Your plan and choose a plan. Billing then opens, where you add a billing email and a payment method.

Security and compliance​

How does Revelry Ai protect our data?​

Revelry Ai does not use your inputs or outputs to train models. It reaches model providers through API agreements that generally prohibit them from training on API data. See the Privacy Policy for details.

Revelry holds a SOC 2 Type 1 report, and the SOC 2 Type 2 audit is in progress.

Data is encrypted in transit with TLS and encrypted at rest, OAuth tokens for connected services are encrypted, and passwords are stored hashed.

When you add content to Knowledge, its text is sent to OpenAI's API (or Azure OpenAI, depending on your deployment) so Knowledge Search can find it by meaning.

How are enterprise deployments set up?​

Enterprise deployments run inside a private VPC with no public access. Enterprise customers can also bring their own model provider accounts: an organization admin adds your API keys on the LLM Providers page, and model requests then run under your organization's provider accounts and agreements. To set this up, email support+enterprise@revelry.ai.

We have HIPAA, FedRAMP, or other compliance requirements. Can Revelry Ai meet them?​

Revelry can deploy Revelry Ai into infrastructure built for HIPAA, FedRAMP, and other compliance requirements. Email support+security@revelry.ai with what you need. The same address handles security questionnaires and any other security questions.

Getting help​

Where can I get help?​

Start with this documentation. If you're still stuck, email support@revelry.ai.

How do I report a bug?​

Email support@revelry.ai with the steps that led to the bug, what you expected to happen, and what happened instead. Screenshots and screen recordings help us reproduce it faster.