--- sourceDocument: Australia Enable AI sourceDocumentLink: https://www.servicenow.com/docs/r/australia/intelligent-experiences Release : - australia ft:locale : - en-US ft:publication_title : - Australia Enable AI ft:clusterId : - platai bundleId : - platai workflow : - Platform --- # Decomposition AI agent # Decomposition AI agent {#ariaid-title1} Release version: Australia Updated August 14, 2026 ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 2 minutes to read Summarize ![AI sparkle icon](https://servicenow.com/docs/portal-asset/ai-sparkle-icon) Summarized using AI This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete. ## Summary of Decomposition AI agent The Decomposition AI agent is designed to enhance resolution workflows by taking existing resolution steps as input and breaking them down into smaller, clear, and actionable substeps. It then creates corresponding records based on user preferences, enabling more precise execution, improved task clarity, and supporting automation and auditability within ServiceNow environments. Show full answer Show less ## Workflow * The agent retrieves table configuration and action policies that specify which tables can be handled autonomously or require supervision. * It obtains resolution steps from memory but does not generate new resolution steps independently. * It decomposes each resolution step into detailed actionable substeps, categorizing them into autonomous and supervised groups. * Autonomous steps proceed automatically, while supervised steps require user input before execution. * Finally, the agent creates the necessary records and saves activity notes, summarizing the execution in the parent record. ## Configuration and Access * **Third-party access:** Can be enabled to allow external AI agents to use this agent (disabled by default). * **AI specialist access:** When enabled, provides additional configuration options for mapping inputs and outputs, allowing specialists more control (disabled by default). * **Long-term memory:** Optionally enabled to use previous user interactions as context for the language model (disabled by default). * These settings are managed through the AI Agent configs table and system properties. ## Security and Permissions * Access is controlled by specific user roles (e.g., `snuxcgenai.platformaidecompositionagent`). * Data access roles define the scope of data the agent can access and the actions it can perform. ## Integration and Usage * The agent supports integration with Virtual Agent and ServiceNow Otto panel via Assistant Designer for conversational workflows. * It can be incorporated into agentic workflows to generate and manage resolution plans efficiently. * Administrators may configure triggers to automate the agent's activation as needed. ## Practical Benefits ServiceNow customers can leverage the Decomposition AI agent to systematically break down complex resolution steps into manageable tasks, improving workflow clarity, enabling partial automation, and maintaining audit trails. This leads to faster, more accurate issue resolution and better governance of resolution processes. This AI agent takes resolution steps as input, and then analyzes and breaks down each resolution step into a clear, structured set of smaller, actionable substeps. It then creates records based on user preference. This enables precise execution, improves task clarity, and supports automation and auditability of resolution workflows. ## Workflow {#plat-ai-decomposition-agent-ai-agent__zsmdlwierhjt} The agent retrieve table configuration, decompose the resolution plan into actionable substeps, create the records and to save the summary in the parent record. It takes resolution steps as input. It does not generate resolution steps on its own. 1. Retrieve the table action policy XML that defines which tables can be acted on autonomously and which require supervision. 2. Retrieve resolution steps from memory. 3. Decompose the steps. 4. Group actionable steps into autonomous and supervised categories. Proceed automatically with autonomous steps. For supervised steps, wait for user input before executing. 5. Finalize the execution set. 6. Create the record. 7. Save activity notes. {#plat-ai-decomposition-agent-ai-agent__ol_hcp_qwx_njc} {#plat-ai-decomposition-agent-ai-agent__zsmdlwierhjt__entry__2} | Field | Description | |-|-| | Allow third party to access this AI agent | When enabled, third-party AI agents can use this agent. This value is off (false) by default. This setting is defined in the AI Agent configs \[sn_aia_agent_config\] table on the External discoverable field. | | Allow AI specialists to access this AI agent | When enabled, AI specialists can use this agent. This value is off (false) by default. When set to true, more configuration options for tools become available so that an AI specialist can map inputs and response templates to tool outputs. This setting is defined in the AI Agent configs \[sn_aia_agent_config\] table on the Specialist enabled field. | | Manage long-term memory | When enabled, all previous user interactions are used as context for the LLM. This value is off (false) by default. This setting is defined by the sn_aia.ltm.enable_long_term_memory system property. For more information, see [ServiceNow Otto AI agents reference](https://www.servicenow.com/docs/MUqSU8rHp3ao~q3QEWgHkA "Find more information about user roles, tables, and the different properties that are installed in ServiceNow Otto AI agents."). | | Tools | Scripts : Create record : Get actionable tables from skill config : Get Fields from Table : Save activity notes Generative AI skill : Step Decomposer | | Allowed user roles The specific user roles that can access this AI agent. | sn_uxc_gen_ai.platform_ai_decomposition_agent | | Data access roles The specific user identity roles that determine which data the AI agent can access and what actions it can take. | sn_uxc_gen_ai.platform_ai_decomposition_agent, sn_hr_core.case_writer | | Triggers | Optional. None defined by default. An admin can specify triggers if desired. For more information, see [Add a trigger to an AI agent](https://www.servicenow.com/docs/NXqRJBQV_Vw00FEzklcAJw "In the guided setup for an AI agent, add triggers to run the AI agent automatically when certain conditions are met."). | | Channels | Configure an assistant for Virtual Agent or ServiceNow Otto panel using [Assistant Designer](https://www.servicenow.com/docs/access?context=configure-now-assist-va&version=australia&pubname=australia-conversational-interfaces&ft:locale=en-US). | | Used in agentic workflows | Generate resolution plans | [Table 1. Configuration]