# Cakewalk for Employees Employees run their Agents through Cakewalk. Connect once per app, then every tool call your Agent makes is governed automatically. Sensitive actions pause inside your Agent client and resume the moment you respond. *** ### ❓ **Why use Cakewalk?** Employees use Cakewalk to run AI Agents safely without changing how they work. It gives you: * **Inline approvals.** Sensitive actions pause inside your Agent client. Approve or deny in place and the Agent resumes with full context. * **One Connection setup per app.** Authorize once. Every Agent you run inherits exactly your access through Cakewalk. * **Per-User scope.** Your Agent's permissions match yours in each app. Each teammate's Agents stay scoped to their own access. * **Personal Agent activity.** You see every session and every tool call your Agent makes on your behalf. * **Credential Mediation.** Your tokens stay in the Cakewalk vault. The gateway injects them at proxy time so your Agent never holds real credentials. *** ### 🛠 **Everyday workflows** #### **Set up your Connections** * Open **My Connections** to see every app your admin has made available. * Click **Connect** on any card. You'll be redirected to the app's sign-in page to authorize. * Once authorized, the card flips to **Connected** and your Agents can access that app through Cakewalk. :bulb: *Why this matters:* Connections are per-User. Your Agent inherits exactly your access in each app. *** #### **Set up your Agent** * Open **My Agents** to see every Agent in the company catalog. * Click **Set up** on the Agent you use. The setup modal walks you through your Agent's specific install method (deep link, JSON config, custom form or CLI). * Once connected, your Agent sends every tool call through Cakewalk. :bulb: *Why this matters:* One setup per Agent. After that, every tool call your Agent makes is governed automatically. *** #### **Approve sensitive actions** * When your Agent tries an action that needs approval, it pauses and shows a prompt inside the Agent client (via MCP elicitation). * Approve or deny in place; your decision lands inside the same Agent client you're already working in. * Auto-approved actions run without prompting. Denied actions stop at the gateway and the downstream app never sees them. :bulb: *Why this matters:* You stay in flow. The Agent waits, then resumes the moment you respond. *** #### **See what your Agents did** * Open **Agents Activity** to see status, last activity and session counts. * Click into any Session for details: which tools were called, what was approved, what was denied. :bulb: *Why this matters:* You always know what your Agent is doing on your behalf. *** #### **Manage your profile** * Open **Profile Settings → Personal Info** to see your team and role. * Log out from the same screen. :bulb: *Why this matters:* Your identity is the source of truth for what your Agents can access. Profile data stays in sync with your HR system. *** ### 👉 **Where to go next** * Get set up in five minutes: [Employee Setup](/docs/ai-agent-access/introduction/get-going-with-agent-access/employee-setup.md). * Understand the model: [The Three Actors](/docs/ai-agent-access/concepts/the-three-actors.md), [The MCP Gateway](/docs/ai-agent-access/concepts/the-mcp-gateway.md). * See the admin side: [Cakewalk for Admins](/docs/ai-agent-access/introduction/cakewalk-for-admins.md). --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://www.cakewalk.security/docs/ai-agent-access/introduction/cakewalk-for-employees.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.