> ## Content Index > Fetch the complete content index at: https://www.stirup.si/llms.txt > Use this file to discover other available public pages before exploring further. # ShipStation Global Integrates Deckhand with Amazon Bedrock AgentCore for AI Help Desk - URL: https://www.stirup.si/shipstation-global-integrates-deckhand-with-amazon-bedrock-agentcore-f-5bfc900b/ - Published: 2026-10-13T16:01:11.000Z - Updated: 2026-10-13T16:01:11.000Z - Description: ShipStation Global's Deckhand AI help desk agent for Slack is built on Amazon Bedrock AgentCore, enabling scalable, secure, and efficient IT support through isolated runtime sessions and modular architecture. - Author: Satish Das - Tags: Cloud, News, Stirup Research ## Overview of Deckhand and Amazon Bedrock AgentCore Integration ShipStation Global's Deckhand is an AI-powered IT help desk agent for Slack, built on Amazon Bedrock AgentCore, as noted by the ShipStation Global Engineering team. The platform, Amazon Bedrock AgentCore, allows for the building, connecting, and optimizing of agents at scale, supporting any framework or model. A key feature of this integration is the AgentCore runtime, which hosts the agent in isolated sessions, eliminating the need for servers or containers. This approach streamlines deployment and enhances security by isolating each session. The integration highlights how ShipStation Global leverages Amazon Bedrock AgentCore to create a scalable and efficient AI help desk solution tailored for internal IT support. ## Architecture and Components of the AI Help Desk System ShipStation Global’s AI IT help desk system is built on Amazon Bedrock AgentCore, leveraging a modular architecture to handle complex user interactions. At its core is a Strands agent, deployed as a container on the AgentCore runtime, which manages multi-turn conversations using Anthropic’s Claude from Amazon Bedrock, as described in [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si). The system integrates tools and knowledge through a single AWS Lambda function, exposed as Model Context Protocol (MCP) tools via the AgentCore Gateway, enabling request classification, Jira ticket creation, and image analysis, as outlined in [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si). The architecture, detailed in Figure 2 ([AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si)), shows Slack events being routed through the Slack service and AgentCore runtime, with separate paths to an Amazon Bedrock knowledge base, AgentCore Gateway-backed Lambda tools, Jira Service Management, and DynamoDB for feedback collection. This design ensures scalability and seamless integration with existing IT workflows. ## Knowledge Base and Data Integration ShipStation Global built its AI IT help desk using Amazon Bedrock AgentCore, which integrates knowledge from multiple internal and external sources. The system pulls content from Confluence, the company intranet, Slack, and marketing pages, storing it in Amazon Simple Storage Service (Amazon S3) for indexing. This centralized knowledge base allows the AI to retrieve and route answers efficiently. According to [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si), the design of the knowledge base ensures that all relevant information is accessible to the AI, improving response accuracy. The pipeline uploads data to Amazon S3, as noted in [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si), ensuring that the knowledge base remains up to date with the latest information from these sources. This integration enables the AI to provide consistent and contextually relevant support to users across the organization. ## Security and Enforcement Mechanisms ShipStation Global implemented strict security and enforcement mechanisms within its AI IT help desk built on Amazon Bedrock AgentCore. According to [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si), ticket creation requires explicit confirmation, ensuring that only authorized requests are processed. [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si) outlines that ticket updates are restricted to valid keys matching the helpdesk project pattern, and empty comments are rejected, reinforcing data integrity. Additionally, [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si) specifies that the image tool is limited to downloading files exclusively from Slack’s file host, minimizing exposure to external threats. These measures collectively enhance the system’s security posture by enforcing strict access controls and validation checks at multiple stages of the help desk workflow. ## Operational Workflow and Service Separation ShipStation Global built its AI IT help desk on Amazon Bedrock AgentCore by separating operational workflows into distinct managed services. AgentCore runtime hosts the agent, while AgentCore Gateway transforms Lambda functions into MCP tools accessible to the agent. Amazon Bedrock Knowledge Bases manages content retrieval, ensuring each component performs a single, well-defined task. Request classification occurs as a separate Amazon Bedrock Converse API call, returning structured ticket fields validated with Pydantic before data is sent to Jira. Developers are directed to the Amazon Bedrock AgentCore documentation and Strands Agents SDK for agent development, and to the Amazon Bedrock Knowledge Bases documentation for connecting content sources. This separation of concerns enables a scalable, modular architecture that aligns with best practices in cloud service design. ## Sources and reporting notes - [AWS Machine Learning](https://aws.amazon.com/blogs/machine-learning/how-shipstation-global-built-an-ai-it-help-desk-on-amazon-bedrock-agentcore/?ref=stirup.si) Based on published source material. No independent hands-on testing is claimed.