We use cookies to enhance your experience and measure how the site performs. Choose "Essential Only" to disable analytics. Read our Privacy Policy.

    Odeus Docs

    Odeus Cheat Sheet

    Your essential companion for getting the most out of Odeus

    Odeus Cheat Sheet

    Your essential companion for getting the most out of Odeus

    Getting Started

    Before we start, thanks for using Odeus!

    If you run into any issues or have ideas for improvements, don't hesitate to reach out to your workspace administrator. Your feedback helps us prioritize what to build next.

    What is Odeus?

    Odeus is your secure, enterprise-ready AI platform that gives your team governed access to multiple Large Language Models — Anthropic Claude, OpenAI GPT, Google Gemini, and Mistral — alongside a Knowledge Hub, Agents, Workflows, and Deep Research. It's designed to help teams leverage AI safely while keeping your data private (your data is never used to train models).

    Accessing Odeus

    Navigate to `https://app.odeus.ai` and sign in with Google SSO or an email magic-link / one-time code.



Type your prompt in the chat interface and press Enter to begin.

    Use Cmd+K (or Ctrl+K on Windows) to open the command palette for quick navigation to previous chats, agents, and settings.

    Model Selection Guide

    Odeus offers models from Anthropic, OpenAI, Google, and Mistral (DeepSeek is coming soon). Choose the right model for your task to balance quality, speed, and cost.

    For the complete list of available models, visit the models page.

    Switch models any time using the model selector in the chat window. You pick the model per conversation, and Odeus will automatically fall back to a capable alternative provider if your chosen model is temporarily unavailable.

    Core Features

    Chat Functionality

    • Web Search — The model can browse the web for fresh information when needed.

    • Share Conversations — Use the share button to share a chat with teammates in your workspace.

    • Voice Input — Dictate your prompt with the microphone and review the transcription before sending.

    • Inline Citations — Answers grounded in your documents include citations down to the file, page, and location.

    Working with Documents

    Click the paperclip icon in the chat input to upload files directly. Supported formats include PDF, DOCX, XLSX, PPTX, CSV, TXT, MD, HTML, and images. PDFs are processed with layout-aware OCR (including Arabic).

    Add documents to the Knowledge Hub for retrieval-augmented answers across your team. You can also connect Google Drive so chosen folders sync into the Knowledge Hub automatically (every ~15 minutes). Other sources such as SharePoint, Slack, and Notion are available as browse/read connectors via MCP rather than full sync.

    • Summarize long documents
    • Extract key information
    • Answer questions with cited sources
    • Compare multiple documents
    • Translate content

    Agents

    Turn repeatable tasks into reusable assistants with custom Agents.

    Creating Agents

    Open **Agents** to browse existing agents or create a new one.



In the agent builder you can:

* Set the instructions / persona
* Pin the model the agent should use
* Attach knowledge sources from the Knowledge Hub
* Add tools (MCP connectors) and Skills



Make the agent available to yourself, your team/workspace, or publish a public no-auth agent page to share a link with anyone.

    Be careful sharing agents that reference sensitive knowledge. Keep them workspace-private unless you intend the content to be broadly accessible.

    Deep Research

    Odeus Deep Research runs a multi-step, multi-source investigation and returns a cited report. Unlike tools that lock the research models, Odeus lets you choose the models per run.

    Select **Deep Research** in the composer and enter your research query.



Choose a depth preset (Quick / Standard / Deep) and the models for the research roles.



Watch the reasoning trace as it works, then review the final report with inline citations.

    Skills & Workflows

    • Skills — Reusable instruction sets (and bundled tools) that you and your workspace can apply across chats and agents.

    • Workflows — Visual, multi-step automations with branching, loops, parallel steps, approvals, and scheduling.

    Security & Privacy

    Odeus is built for enterprise use:

    • Your data is never used to train or evaluate LLMs
    • Multi-tenant orgs with role-based access (owner / admin / billing / member)
    • Google SSO and email magic-link sign-in
    • Per-agent and per-workspace access controls

    Enterprise governance such as SAML, SCIM, audit logs, and custom data-retention policies is on the roadmap (coming soon).

    Best Practices

    • Use Odeus, not direct LLM access — Use Odeus instead of consumer LLM apps for company work, so usage stays governed and private.

    • Share thoughtfully — Use workspace and team scopes to keep sensitive agents and knowledge with the right people.

    Getting Help

    • Prompting basics — Templates and best practices for getting better results.

    • Feedback — Send suggestions to your workspace administrator.