# ManageEngine ServiceDesk Plus AI Agent The **ServiceDesk Plus AI Agent** enables end users to interact with their IT service desk through natural language, while executing real actions directly in **ManageEngine ServiceDesk Plus (SDP)** via Autom Mate. *** ### Overview The agent is designed to handle the most common support interactions without human intervention: * Create new requests * View user’s open requests * Add notes to existing requests * Search knowledge base articles All actions are executed through **Autom Mate workflows**, ensuring full auditability, control, and integration with existing SDP processes. *** ### Prerequisites #### Autom Mate * Access to **AI Agent Composer**\ 👉 [AI Agent Composer](https://docs.autommate.com/userguide/ai-agent-composer) *** #### AI Model The agent requires one of the following providers: * **OpenAI** * **Azure OpenAI** Credentials must be configured in Autom Mate Vault: 👉 [Vault](https://docs.autommate.com/userguide/product-features/management/vault) Credential configuration includes: * API Key * Endpoint (for Azure OpenAI) * Model / Deployment selection *** #### ManageEngine ServiceDesk Plus You must provide: * **SDP API access** * **Authentication** Supported methods: * API Key Credentials must be configured in Autom Mate Vault. *** ### Required Permissions At minimum, the integration user should be able to: * Create requests * Read requests * Add notes to requests * Read knowledge base articles If the design requires the agent to return only the logged-in user’s own requests, identity mapping between the channel user and the corresponding ServiceDesk Plus user must also be available. *** ### Knowledge Integration Knowledge article support can be implemented in two ways depending on the environment design. #### Option 1: API-based knowledge lookup The agent queries ServiceDesk Plus knowledge records directly during execution. #### Option 2: RAG-based knowledge retrieval Knowledge articles are ingested into the agent’s knowledge layer through Autom Mate, allowing the agent to retrieve and rank relevant content before responding. For this use case, the knowledge capability is limited to: * Searching for relevant articles * Returning the most relevant matches * Suggesting existing guidance before creating a new request This allows the agent to reduce unnecessary ticket creation for known issues or common how-to questions. *** ### Capabilities (Out-of-the-Box) This agent supports the following actions out of the box. #### Create Request The agent can create a new request in ServiceDesk Plus based on the user’s description. **Example** > My laptop is not connecting to Wi-Fi. **What happens** * The agent extracts the issue details * Autom Mate maps the input to the request payload * A request is created in ServiceDesk Plus * The agent returns the created request number or confirmation #### View User’s Open Requests The agent can return the current user’s open requests. **Example** > Show my open tickets. **What happens** * The user is identified * Requests are fetched from ServiceDesk Plus * Open or active requests belonging to that user are returned * The result is formatted into a simple list #### Add Note to Request The agent can append a note to an existing request. **Example** > Add a note to ticket 143 saying the issue only happens on VPN. **What happens** * The request ID is validated * The note is passed to ServiceDesk Plus * The note is added to the selected request * The agent confirms the update #### Search Knowledge Articles The agent can search for relevant knowledge articles based on the user’s question. **Example** > How do I reset my VPN password? **What happens** * The agent searches the configured knowledge source * Relevant articles are retrieved * The response is returned as suggested guidance * The user may resolve the issue without creating a ticket *** ### Execution Flow The agent follows a simple request-response pattern, with Autom Mate handling all execution logic. #### Example: Create Request ``` User Message → Intent Detection → Extract Request Details → Create Request in ServiceDesk Plus → Return Confirmation ``` #### Example: View Open Requests ``` User Message → Identify User → Fetch Requests → Filter User's Open Requests → Return Result ``` #### Example: Add Note ``` User Message → Extract Request ID and Note Content → Validate Request → Add Note → Return Confirmation ``` #### Example: Knowledge Search ``` User Message → Search Knowledge Source → Rank Relevant Articles → Return Suggested Articles ``` *** ### Execution Model * The agent follows the standard Autom Mate execution model: ``` User → AI Agent → Autom Workflow → ServiceDesk Plus ``` #### * #### AI Agent The AI Agent handles conversation, intent understanding, and response orchestration. #### * #### Autom Workflow The Autom workflow acts as the execution layer. It performs validation, mapping, filtering, and API interaction. #### * #### ServiceDesk Plus ServiceDesk Plus remains the system of record for requests, notes, users, and knowledge content. This model keeps the agent lightweight while ensuring all business actions remain visible and controlled through Autom Mate. *** ### Safety and Permissions The agent should operate with clearly scoped capabilities. Recommended controls include: * Restricting the integration account to only the required actions * Returning only the user’s own requests * Validating request IDs before note updates * Logging all executions through Autom Mate * Limiting the agent to read and write actions that are explicitly supported For this use case, the agent is intentionally narrow in scope. It does not perform administrative actions or broader system-level operations. *** ### Extending the Agent This agent can be extended later with additional ServiceDesk Plus capabilities depending on project scope. Possible extensions include: * Request detail retrieval * Request status updates * Approval interactions * Technician-specific actions * User lookup and enrichment * Additional knowledge actions * Multi-step request guidance The current version is intentionally focused on a small set of stable, high-frequency use cases that are easy to deploy and validate. 👉 [AI Agent Composer](https://docs.autommate.com/userguide/ai-agent-composer) *** ### Channels The agent can be exposed through any supported Autom Mate channel, including: * Web chat * Microsoft Teams * WhatsApp * Other supported conversational channels The execution logic remains the same across channels. Only the interaction layer changes. *** ### Summary The **ManageEngine ServiceDesk Plus AI Agent** provides a practical self-service layer for common support interactions. It allows users to: * Create requests * View their own open requests * Add notes to existing requests * Search knowledge articles All actions are executed through **Autom Mate workflows**, giving the organization a controlled and auditable way to deliver AI-powered support on top of ServiceDesk Plus. --- # 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://docs.autommate.com/userguide/use-cases/ai-agent-use-cases/manageengine-servicedesk-plus-ai-agent.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.