{"revision":"c10276e953456835cdbb1f3a60d58ae1980be353","document":{"id":23428,"path":"intelligent-experiences/gov-sec-review-kill-switch-protocol-log.md","title":"Review the agent containment list","bundle":null,"doc_type":null,"release":"australia","body":"\n# Review the agent containment list\n\nThe agent containment list shows agents that were deactivated and reinstated for the instance manually with kill switch protocol or automatically by policy enforcement. You can view audit log information for each AI agent which can help you stay compliant with regulatory guidance and your business rules.\n\n## Before you begin\n\nRole required: AI steward \\[sn\\_ai\\_governance.ai\\_steward\\]\n\n## Procedure\n\n1.  Navigate to **Security** &gt; **Overview** &gt; **Contained AI Agents**.\n\n<table><thead><tr><th>\n\nField\n\n</th><th>\n\nDescription\n\n</th></tr></thead><tbody><tr><td>\n\nStatus\n\n</td><td>\n\nThe current state of the operation. Possible values are:-   Completed — The operation was successful. If the agent was deactivated, all AI agent credentials were revoked across all providers, the session was ended, and no new access tokens will be provisioned. If the agent was reinstated, all AI agent credentials were restored across all providers.\n-   Failed — The containment or reinstatement operation didn't take effect on any provider. Failed to deactivate or reinstate agent on all configured providers or skipped providers \\(for example, agent not found, subflow error, or an unhandled exception\\).\n-   In progress — The operation is actively executing. Policy enforcement point \\(PEP\\) subflows \\(for example, AWS Bedrock\\) are running, with audit log information being captured.\n-   Partial — The operation succeeded on at least one provider and failed on at least one provider. For example, the AWS Bedrock AI agent deactivation succeeded, but the Okta deactivation failed.\n\n\n</td></tr><tr><td>\n\nAI agent name\n\n</td><td>\n\nThe name of the contained AI agent. Select the name to go to the AI asset in Inventory.\n\n</td></tr><tr><td>\n\nStart time\n\n</td><td>\n\nThe date and time when the operation was initiated.\n\n</td></tr><tr><td>\n\nEnd time\n\n</td><td>\n\nThe date and time when the operation ended, regardless of status.\n\n</td></tr><tr><td>\n\nDomain\n\n</td><td>\n\nThe domain the AI agent belongs to. If global or null is shown, the instance isn't domain-separated.\n\n</td></tr><tr><td>\n\nOperation\n\n</td><td>\n\nThe operation that was attempted. Possible values are:-   Deactivate\n-   Reinstate\n\n\n</td></tr><tr><td>\n\nActions\n\n</td><td>\n\nSelect one of the following:-   **View details** — Opens a side panel with two subtabs: **Overview** and **Audit log**. The Overview subtab shows containment context \\(including trigger type of Manual or Automated\\) and identity and enforcement information.\n-   **Reinstate agent** — Make the AI agent active again.\n\n\n</td></tr></tbody>\n</table>2.  Select **All**, **In progress**, or **Contained** to filter the list.\n\n\n**Parent Topic:**[Contain AI agents manually using kill switch protocol](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/australia/markdown/australia/intelligent-experiences/gov-sec-manage-ai-agents-using-kill-switch-protocol.md)\n\n"}}