使用 Amazon Quick Desktop 和 Amazon FSx for NetApp ONTAP 构建受治理的报告流程

Amazon Quick Desktop 将受治理、由 AI 辅助的报告能力带到团队已在 Amazon FSx for NetApp ONTAP（FSx for ONTAP）中管理的文件中，可将每周报告的准备时间从数小时缩短至数分钟。如今，制作这些报告每周仍需要数小时的手动工作。团队需要反复阅读相同文档、重新格式化指标，并将摘要复制到 Slack 中。管理者需要快速获得这些更新，但他们同样需要能够信任每项指标、风险和建议背后的来源。

在本文中，您将使用 Amazon Quick Desktop、Amazon Quick skills、Amazon FSx for NetApp ONTAP、Amazon Simple Storage Service（Amazon S3）access point 和 Slack，构建一个受治理的每周报告工作流。Amazon Quick Desktop 是一个原生桌面应用程序，可将 Amazon Quick 从浏览器扩展到用户的计算机。它提供 Web 体验中的 AI 驱动能力，以及以下桌面专属功能：

• 本地文件访问。
• 后台处理。
• 个人知识图谱。

在 Amazon Quick Desktop 中，您将创建一个名为 Weekly Business Reporting Assistant 的 skill。Skill 是一种可复用工作流，可为 Amazon Quick 提供特定任务的指令，使用户能够每周通过一个触发短语启动相同的报告流程。

该工作流将一组受治理的业务文件连接到引导式报告体验中。它让源文档保留在现有存储和访问控制体系之下，仅允许 Amazon Quick 访问选定的报告内容，并且在共享工件或 Slack 摘要之前保留人工审核环节。示例使用虚构的业务数据，但这一模式同样适用于运营审查、预测更新、风险报告、高管简报和每周业务审查。

解决方案概述

该解决方案将 Amazon Quick Desktop 作为每周报告的主要工作空间。业务用户打开 Quick Desktop，激活 Weekly Business Reporting Assistant skill，并提出诸如“自上周运营审查以来发生了哪些变化？”或“为管理层创建一份单页周报”等问题。

该 skill 使工作流可重复执行。它向 Amazon Quick 提供覆盖源内容收集、输出格式、引用行为以及何时要求澄清的特定任务指令。这使用户能够从引导式报告流程开始，而非面对一个空白对话。

源文件仍保留在 FSx for ONTAP 中。在本演练中，FSx for ONTAP 存储支持每周报告的文档，例如最终版周度审查、运营计划、预测摘要、风险登记册和高管模板。将文件保留在 FSx for ONTAP 中，可让存储团队和业务团队继续使用既有的文件访问与治理流程。

您为该助手选择一个报告文件夹。该文件夹包含业务所有者为该工作流批准的文件。草稿、过期文件、受限文档和无关内容均保留在该文件夹之外。

S3 access point 向 Amazon Quick 提供对这一选定文件夹的受控只读访问。AWS Identity and Access Management（IAM） 权限会限制 Amazon Quick 可以列出和读取的内容。

Amazon Quick 使用选定文件创建一个可搜索的源内容集合，称为 knowledge base。在本演练中，该 knowledge base 名为 Business Reporting Archive。Weekly Business Reporting Assistant skill 使用它回答问题、引用源文档、创建报告、生成演示文稿和 PDF、构建可视化内容，以及起草可直接用于 Slack 的摘要。

报告准备好分享后，您可以使用 Amazon Quick 中的 Slack 集成起草频道摘要。您需要在发布消息之前对其进行审核和批准。

图 1 展示了该工作流。业务文件保留在 FSx for ONTAP 中。S3 access point 向 Amazon Quick 提供对选定报告文件夹的受控访问。Amazon Quick 将该内容准备为 Business Reporting Archive knowledge base。用户在 Quick Desktop 中使用 Weekly Business Reporting Assistant skill 创建带引用的报告工件和 Slack 摘要，供审批后使用。

图 1：使用 Amazon Quick Desktop、Amazon Quick skills、Amazon FSx for NetApp ONTAP、Amazon S3 access point 和 Slack 的受治理每周报告工作流

Amazon Quick Desktop 还会在 My context 中构建个人知识图谱，连接人员、项目、事件、行动事项和文档。例如，在准备每周报告时，用户可以从已连接的来源中识别相关人员、行动事项、文档和 Slack 频道。Business Reporting Archive 仍然是报告事实和引用内容的受治理来源。

图 2：Amazon Quick Desktop 的 My context 视图中的个人知识图谱示例

前提条件

本演练假设您的存储和身份基础设施已经存在。您将配置访问路径、Amazon Quick knowledge base、报告助手、Slack 操作连接器和验证工作流。您无需创建 FSx for ONTAP 文件系统本身。

AWS 账户和 Amazon Quick 要求

• 一个 AWS 账户，且可访问 Amazon Quick、Amazon FSx for NetApp ONTAP 和适用于 FSx for ONTAP 的 Amazon S3 access point 所支持的 Region。
• 一个 Amazon Quick Enterprise 订阅，并具有创建 Amazon S3 集成、knowledge base 和 space 的权限。
• 为运行报告工作流的用户安装 Amazon Quick 桌面应用程序。在生产环境推广之前，请审查企业部署要求。

FSx for ONTAP 要求

• 一个运行 NetApp ONTAP 9.17.1 或更高版本的 FSx for ONTAP 文件系统。
• 一个已挂载且具有 junction path 的 FSx for ONTAP 卷，例如 /business-reporting。
• 一个附加到 FSx for ONTAP 卷的 S3 access point，或创建该 access point 的权限。S3 access point 和 FSx for ONTAP 卷必须位于同一 AWS Region，并由同一个 AWS 账户拥有。
• 授予 Amazon Quick 对 S3 access point 和已批准对象前缀最小权限读取访问权限的 AWS Identity and Access Management（IAM）权限。
• 与您的 Amazon Quick 应用程序位于同一 AWS Region 的 Amazon S3 路径。

Slack 和测试要求

• 一个 Slack 工作区、一个用于已审核摘要的获批准频道，以及配置 Slack 操作连接器的权限。
• 一组用于测试的、范围聚焦的虚构或已批准业务文件，例如周度审查、运营计划、预测摘要、风险登记册和高管模板。

实施

实施过程包括七个步骤。以下章节将逐一介绍：准备源文件夹、创建 S3 access point、授予 Amazon Quick 读取访问权限、创建 Amazon S3 集成和 knowledge base、配置 Slack、创建报告助手，以及测试工作流。

第 1 步：准备源内容

首先限定较窄的文件夹范围，以便业务用户能够根据一组已知源文件验证引用。对于本演练，请使用一个已挂载的 FSx for ONTAP 卷，并创建或选择一个已批准的文件夹，例如 /business-reporting。

在该文件夹内，围绕报告流程组织文件。添加最近 8–12 份每周业务审查、当前运营计划、当前预测摘要和当前风险登记册。还应添加用于指导输出风格的标准报告模板。在建立索引前，移除草稿、过时、归档和受限文件，并确认文件所有者同意该源内容范围。

/business-reporting/
    weekly-reviews/
    operating-plans/
    forecasts/
    risk-registers/
    executive-templates/

图 3：挂载的 FSx for ONTAP 卷上已批准报告文件夹结构示例

第 2 步：创建 S3 access point

将 S3 access point 附加到 FSx for ONTAP 卷。随后，Amazon Quick 可通过兼容 S3 的路径读取已批准内容。如果您不拥有 FSx for ONTAP 配置，请与您的存储管理员或云管理员协作。

1. 在包含 FSx for ONTAP 卷的 Region 中打开 Amazon FSx 控制台。
2. 在导航窗格中，选择 Volumes。
3. 选择包含已批准报告文件夹的、已挂载的 FSx for ONTAP 卷。
4. 在 Actions 中，选择 Create S3 access point。
5. 输入符合您命名标准的 access point 名称，例如 business-reporting-ap。
6. 根据您的治理模型配置 access point 用户和网络设置。如果工作负载需要私有网络访问，请选择虚拟私有云限制。
7. 创建 access point，然后记录 access point 别名和 Amazon Resource Name（ARN）。在 S3 URI 中使用该别名，并在 IAM 策略中使用 ARN。

图 4：FSx for ONTAP 数据源的 S3 access point 概览

第 3 步：授予 Amazon Quick 读取访问权限

仅授予 Amazon Quick 列出和读取已批准前缀所需的权限。以下策略可作为起点。请将占位符替换为您的 Region、账户 ID、access point 名称和前缀。

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "ListApprovedReportingPrefix",
            "Effect": "Allow",
            "Action": ["s3:ListBucket", "s3:GetBucketLocation"],
            "Resource": "<access-point-arn>"
        },
        {
            "Sid": "ReadApprovedReportingObjects",
            "Effect": "Allow",
            "Action": "s3:GetObject",
            "Resource": "<approved-prefix-object-arn>"
        }
    ]
}

将 <access-point-arn> 设置为 arn:aws:s3:<region>:<account-id>:accesspoint/<access-point-name>。将 <approved-prefix-object-arn> 设置为 arn:aws:s3:<region>:<account-id>:accesspoint/<access-point-name>/object/business-reporting/*。

识别 Amazon Quick 在您的账户中用于 Amazon S3 集成的 IAM role 或 service role。将该策略附加到该 role，并授予对 Amazon S3 集成所需的文档级访问控制列表（ACL）或元数据文件的读取权限。

在配置 Amazon S3 集成之前，请验证 access point 别名和前缀。确认该 role 无法读取已批准报告前缀之外的文件夹。

图 5：包含 S3 access point 权限的 IAM 策略编辑器示例

第 4 步：创建 Amazon S3 集成

在 Amazon Quick 中创建 Amazon S3 集成，并使用它根据已批准前缀创建 knowledge base。即使您使用的是 S3 access point 别名，控制台也可能将该路径标记为 bucket URL。在发布助手前，请在您的账户中验证该别名路径。

1. 打开 Amazon Quick 控制台。
2. 选择 Knowledge。
3. 在 Amazon S3 下，选择 Add。
4. 选择现有 Amazon S3 数据源，或选择 + Add account 以连接新数据源。
5. 对于 Name，输入 FSx ONTAP Business Reports。对于 S3 location，输入 access point 别名路径，例如 s3:///amzn-s3-demo-business-reporting/。
6. 在 knowledge base 页面中，输入 FSx ONTAP Business Reports 作为 knowledge base 名称。仅选择已批准的前缀。
7. knowledge base 同步完成后，在 Amazon Quick 控制台中打开 Spaces。
8. 创建一个名为 Business Reporting Archive 的新 space。
9. 将 FSx ONTAP Business Reports knowledge base 添加到该 space。

如果您的报告文件要求按用户或按组设置权限，请先配置文档级 ACL 和元数据设置。Amazon Quick 要求您在创建 knowledge base 时作出 ACL 决策。选择 Create 后，请等待首次同步完成。

（原文在此处截断。）

Amazon Quick Desktop brings governed, AI-assisted reporting to the files your team already manages on Amazon FSx for NetApp ONTAP (FSx for ONTAP), cutting weekly report preparation from hours to minutes. Today, producing those reports takes hours of manual effort each week. Teams re-read the same documents, reformat metrics, and copy summaries into Slack. Leaders need those updates quickly, but they also need to trust the source behind each metric, risk, and recommendation.

In this post, you build a governed weekly reporting workflow with Amazon Quick Desktop, Amazon Quick skills, Amazon FSx for NetApp ONTAP, Amazon Simple Storage Service (Amazon S3) access points, and Slack. Amazon Quick Desktop is a native desktop application that extends Amazon Quick from the browser to the user’s computer. It provides the AI-powered capabilities of the web experience together with desktop-specific capabilities such as

• Local file access.
• Background processing.
• A personal knowledge graph.

In Amazon Quick Desktop, you create a skill called Weekly Business Reporting Assistant. A skill is a reusable workflow that gives Amazon Quick task-specific instructions, so users can start the same reporting process each week with a trigger phrase.

The workflow connects a governed set of business files to a guided reporting experience. It keeps source documents under existing storage and access controls, gives Amazon Quick access only to the selected reporting content, and keeps human review in place before artifacts or Slack summaries are shared. The examples use fictitious business data, but the pattern applies to operating reviews, forecast updates, risk reporting, executive briefings, and weekly business reviews.

Solution overview

This solution makes Amazon Quick Desktop the main workspace for weekly reporting. Business users open Quick Desktop, activate the Weekly Business Reporting Assistant skill, and ask questions such as “What changed since last week’s operating review?” or “Create a one-page weekly report for leadership.”

The skill makes the workflow repeatable. It gives Amazon Quick task-specific instructions that cover the source collection, output formatting, citation behavior, and when to ask for clarification. This lets users start from a guided reporting process instead of a blank conversation.

The source files remain in FSx for ONTAP. In this walkthrough, FSx for ONTAP stores the documents that support weekly reporting, such as final weekly reviews, operating plans, forecast summaries, risk registers, and executive templates. Keeping the files in FSx for ONTAP lets storage and business teams continue using their existing file access and governance processes.

You choose one reporting folder for the assistant to use. This folder contains the files business owners approve for the workflow. Drafts, outdated files, restricted documents, and unrelated content stay outside this folder.

The S3 access point gives Amazon Quick controlled read access to that selected folder. AWS Identity and Access Management (IAM) permissions limit what Amazon Quick can list and read.

Amazon Quick uses the selected files to create a searchable source collection called a knowledge base. In this walkthrough, the knowledge base is named Business Reporting Archive. The Weekly Business Reporting Assistant skill uses it to answer questions, cite source documents, create reports, generate decks and PDFs, build visuals, and draft Slack-ready summaries.

After the report is ready to share, you use Slack integration in Amazon Quick to draft a channel summary. You review and approve the message before posting.

Figure 1 shows the workflow. Business files stay in FSx for ONTAP. The S3 access point gives Amazon Quick controlled access to the selected reporting folder. Amazon Quick prepares that content as the Business Reporting Archive knowledge base. Users work in Quick Desktop with the Weekly Business Reporting Assistant skill to create cited reporting artifacts and Slack summaries for approval.

Figure 1: Governed weekly reporting workflow with Amazon Quick Desktop, Amazon Quick skills, Amazon FSx for NetApp ONTAP, Amazon S3 access points, and Slack

Amazon Quick Desktop also builds a personal knowledge graph in My context , connecting people, projects, events, actions, and documents. For example, while preparing a weekly report, a user can identify related people, action items, documents, and Slack channels from connected sources. The Business Reporting Archive remains the governed source for report facts and citations.

Figure 2: Example personal knowledge graph in the My context view of Amazon Quick Desktop

Prerequisites

This walkthrough assumes your storage and identity foundation already exists. You configure the access path, Amazon Quick knowledge base, reporting assistant, Slack action connector, and validation workflow. You don’t create the FSx for ONTAP file system itself.

AWS account and Amazon Quick requirements

• An AWS account with access to a Region where Amazon Quick, Amazon FSx for NetApp ONTAP, and Amazon S3 access points for FSx for ONTAP are supported.
• An Amazon Quick Enterprise subscription with permissions to create Amazon S3 integrations, knowledge bases, and spaces.
• The Amazon Quick desktop application installed for the users who run the reporting workflow. Review enterprise setup requirements before a production rollout.

FSx for ONTAP requirements

• An FSx for ONTAP file system running NetApp ONTAP 9.17.1 or later.
• An FSx for ONTAP volume that is mounted and has a junction path, such as /business-reporting.
• An S3 access point attached to the FSx for ONTAP volume, or permissions to create one. The S3 access point and the FSx for ONTAP volume must be in the same AWS Region and owned by the same AWS account.
• AWS Identity and Access Management (IAM) permissions to grant Amazon Quick least-privilege read access to the S3 access point and approved object prefix.
• An Amazon S3 path in the same AWS Region as your Amazon Quick application.

Slack and testing requirements

• A Slack workspace, an approved channel for reviewed summaries, and permissions to configure the Slack action connector.
• A focused set of fictitious or approved business files for testing, such as weekly reviews, operating plans, forecast summaries, risk registers, and executive templates.

Implementation

The implementation has seven steps. The following sections walk through each one: prepare the source folder, create the S3 access point, grant Amazon Quick read access, create the Amazon S3 integration and knowledge base, configure Slack, create the reporting assistant, and test the workflow.

Step 1: Prepare source content

Start with a narrow folder scope so business users can verify citations against a known set of source files. For this walkthrough, use a mounted FSx for ONTAP volume and create or choose an approved folder such as /business-reporting.

Within that folder, organize files around the reporting process. Add the last 8–12 weekly business reviews, the current operating plan, the current forecast summary, and the current risk register. Also add the standard report template that guides the output style. Remove draft, obsolete, archived, and restricted files before indexing, and confirm that file owners agree with the source scope.

/business-reporting/
    weekly-reviews/
    operating-plans/
    forecasts/
    risk-registers/
    executive-templates/

Figure 3: Example approved reporting folder structure on a mounted FSx for ONTAP volume

Step 2: Create an S3 access point

Attach an S3 access point to the FSx for ONTAP volume. Amazon Quick can then read the approved content through an S3-compatible path. Work with your storage or cloud administrator if you don’t own the FSx for ONTAP configuration.

1. Open the Amazon FSx console in the Region that contains the FSx for ONTAP volume.
2. In the navigation pane, choose Volumes.
3. Select the mounted FSx for ONTAP volume that contains the approved reporting folder.
4. From Actions , choose Create S3 access point.
5. Enter an access point name that follows your naming standard, such as business-reporting-ap.
6. Configure the access point user and network settings according to your governance model. Select a virtual private cloud restriction if your workload requires private network access.
7. Create the access point, and then record the access point alias and Amazon Resource Name (ARN). Use the alias in the S3 URI and the ARN in IAM policies.

Figure 4: S3 access point overview for an FSx for ONTAP data source

Step 3: Grant Amazon Quick read access

Grant only the access Amazon Quick requires to list and read the approved prefix. The following policy is a starting point. Replace the placeholders with your Region, account ID, access point name, and prefix.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "ListApprovedReportingPrefix",
            "Effect": "Allow",
            "Action": ["s3:ListBucket", "s3:GetBucketLocation"],
            "Resource": "<access-point-arn>"
        },
        {
            "Sid": "ReadApprovedReportingObjects",
            "Effect": "Allow",
            "Action": "s3:GetObject",
            "Resource": "<approved-prefix-object-arn>"
        }
    ]
}

Set <access-point-arn> to arn:aws:s3:<region>:<account-id>:accesspoint/<access-point-name>. Set <approved-prefix-object-arn> to arn:aws:s3:<region>:<account-id>:accesspoint/<access-point-name>/object/business-reporting/*.

Identify the IAM role or service role that Amazon Quick uses for the Amazon S3 integration in your account. Attach the policy to that role, and grant read access to document-level access control list (ACL) or metadata files that the Amazon S3 integration requires.

Before you configure the Amazon S3 integration, validate the access point alias and prefix. Verify that the role cannot read folders outside the approved reporting prefix.

Figure 5: Example IAM policy editor with S3 access point permissions

Step 4: Create the Amazon S3 integration

Create an Amazon S3 integration in Amazon Quick and use it to create a knowledge base from the approved prefix. The console might label the path as a bucket URL even when you use an S3 access point alias. Validate the alias path in your account before you publish the assistant.

1. Open the Amazon Quick console.
2. Choose Knowledge.
3. Under Amazon S3 , choose Add.
4. Choose an existing Amazon S3 data source, or choose + Add account to connect a new source.
5. For Name , enter FSx ONTAP Business Reports. For the S3 location, enter the access point alias path, such as s3:///amzn-s3-demo-business-reporting/.
6. On the knowledge base page, enter FSx ONTAP Business Reports as the knowledge base name. Select only the approved prefix.
7. After the knowledge base sync, open Spaces in the Amazon Quick console.
8. Create a new space called Business Reporting Archive.
9. Add the FSx ONTAP Business Reports knowledge base to this space.

If your reporting files require per-user or per-group permissions, configure document-level ACL and metadata settings first. Amazon Quick requires you to make the ACL decision during knowledge base creation. After you choose Create , wait for the first sync to finish.

Figure 6: Amazon Quick knowledge base created from the approved S3 access point path

Step 5: Configure Slack

Configure Slack so the assistant can stage business summaries for review and posting to the approved channel. Choose the authentication method that matches your organization’s security requirements.

1. In the Amazon Quick console, choose Connectors.
2. In the Amazon Quick desktop application, choose Settings , choose Capabilities , choose Connections , and then choose Browse Connections. This opens the Amazon Quick web console to complete the connection.
3. On the Available tab, find Slack and choose Connect. For a team-managed app or token-based setup, use the Create for your team tab.
4. Complete the Slack sign-in or token flow and grant only the required permissions. Use the minimum required Slack OAuth scopes (for example, chat:write scoped to the approved channel), and rotate credentials on a defined schedule.
5. If your organization requires connector approval, work with your Amazon Quick administrator to publish the Slack connector to your user or group.
6. Confirm the approved Slack channel, such as #finance-weekly, and document who can approve messages before posting.
7. Return to the desktop application and test with a draft message in a non-production channel before using Slack for weekly reporting.

Step 6: Create the reporting assistant

Create a custom skill in the Amazon Quick desktop application that loads the Business Reporting Archive space and applies reporting guardrails. The skill provides a reusable assistant that users activate with a trigger phrase. For more on skill authoring, refer to Amazon Quick skill authoring documentation.

1. In the Amazon Quick desktop application, open a conversation and enter a condensed skill specification like the following.

   Save this as a skill
   
   Skill: "Weekly Business Reporting Assistant"
   
   Trigger: weekly business reviews, operating plans, financial reports, WBR summaries, trend analysis, business reporting deliverables from the Business Reporting Archive
   
   Depends on: quick_suite__spaces
   
   Workflow:
   
   1. [Deterministic] Load Space Context --- search_spaces("Business Reporting Archive") -> list_space_documents(space_id) to get full document list. On failure: "I couldn't load the Business Reporting Archive. Please check that the space is connected and accessible."
   2. [Agentic] Greet and Suggest --- Show greeting with 4 example questions:
      - Top 5 changes since last week's operating review
      - Risks repeating across last 4 WBRs
      - One-page weekly report (highlights, risks, action items, asks, key metrics, decisions needed)
      - Six-slide PowerPoint deck for leadership from latest review + operating plan
   3. [Agentic] Answer Questions --- Use query_topic or search_relevant_content to find relevant content. Respond using structured format: Executive Summary -> Key Highlights -> Risks/Issues -> Action Items/Asks -> Sources (with doc names + dates cited).
   4. [Agentic] Generate Deliverables (if requested) --- Reports structured as: business unit, period covered, key metrics/highlights, risks/blockers, mitigation plans, requests/escalations.
   
   Response tone: Professional business. Always cite specific document name, section, and date.
   
   Safety: Only reference content from the archive. Never fabricate metrics. State gaps clearly.
   
   Lessons Learned:
   - Always load space context before greeting
   - Always cite doc name + date
   - Don't extrapolate metrics not in source docs
   - Ask user when period is ambiguous, business unit unclear, or audience/format not specified
   - If a metric isn't tracked, name what IS available
   

2. Name the skill Weekly Business Reporting Assistant and set a trigger phrase, such as weekly business report.

3. Save the skill and confirm that it loads the Business Reporting Archive space.

Figure 7: Entering the reporting assistant skill specification in Amazon Quick Desktop

Figure 8: Amazon Quick suggests skill defaults before saving the assistant

Step 7: Use Amazon Quick Desktop as the reporting workspace

After you create the skill, activate it from a conversation in the Amazon Quick desktop application. The following example shows a business user moving from a question to a cited report and reviewed Slack summary.

1. Start a new conversation and enter the trigger phrase, such as I need help with business reporting.
2. The skill loads the Business Reporting Archive space and greets you with example questions.

Figure 9: Reporting assistant activated in Amazon Quick Desktop

3. Enter a reporting request, such as What changed since the prior operating review?

Figure 10: Week-over-week change analysis generated from the Business Reporting Archive

Review the cited findings before you use the response. Amazon Quick Desktop shows source details and response steps so you can check the documents and dates behind the response.

Figure 11: Cited key developments and action items from the weekly reporting response

Figure 12: Response steps and source details for review to review before using the response

4. Ask Amazon Quick to create a weekly report, PowerPoint deck, PDF, or executive visual from the cited insights.
5. Review the proposed plan, suggest changes if needed, and choose Approve & Build when the plan is ready.

Figure 13: Report creation plan staged for review before Amazon Quick builds the artifact

6. After you approve the plan, Amazon Quick creates the asset in the requested format.

Figure 14: Generated weekly report artifact in Amazon Quick Desktop

7. Ask Amazon Quick Desktop to draft a Slack summary for the approved channel.
8. Review the draft in the conversation and confirm only when it is ready to post.

Figure 15: Slack-ready reporting summary staged for approval in Amazon Quick Desktop

Figure 16: Approved weekly reporting summary posted to the Slack test channel

Test the workflow with business prompts

After you save the skill, validate the end-to-end workflow with prompts that match real weekly reporting work. Each response must cite specific source documents, include a source list in generated artifacts, and ask for your confirmation before posting to Slack. The following table shows representative prompts and what to look for in the response.

Test area	Prompt	Expected result
Research	What are the top five changes since the prior operating review?	A cited summary with source documents and dates.
Risk analysis	What risks repeat across the last four weekly business reviews?	Grouped risks with citations and suggested owners when the source documents include them.
Report creation	Create a one-page weekly report with highlights, risks, action items, asks, key metrics, and decisions needed.	A structured report with cited claims and a source list.
Deck creation	Create a six-slide PowerPoint deck for leadership from the latest review and operating plan.	A deck or deck outline with title, summary, highlights, risks, decisions, and sources.
Slack distribution	Create a six-bullet Slack summary and ask me before posting.	A Slack-ready message staged for approval before posting.

Security and governance best practices

Move from the sample workflow to a shared reporting workflow in stages. Start with one approved folder path and expand after business users validate citation quality. Keep source file permissions aligned with your existing FSx for ONTAP governance model, and use least-privilege IAM permissions for the role that Amazon Quick uses to read the S3 access point.

Use document-level ACLs in the Amazon S3 integration when the knowledge base requires per-user or per-group access controls. Use a dedicated Slack channel for reporting summaries, define who can approve posts, and require human review for board materials, regulatory content, investor communications, or sensitive financial reporting. Review indexed source files on a schedule and remove obsolete content from the approved folder path.

Troubleshooting

Refer to the following table to diagnose common setup and runtime issues.

Issue	What to check
Amazon Quick cannot read the source path	Confirm the S3 access point alias, same-Region requirement, IAM policy resource scope, and Amazon Quick service role permissions.
Knowledge base sync fails	Confirm that the approved prefix exists, the access point is attached to a mounted FSx for ONTAP volume, and the file system is running NetApp ONTAP 9.17.1 or later.
Responses do not include citations	Confirm that the Business Reporting Archive space is connected in Amazon Quick Desktop and that the knowledge base files synced successfully.
Responses refer to old files	Confirm the knowledge base sync status and remove obsolete files from the approved folder path.
Slack posting fails	Confirm the Slack connector authorization, channel access, required scopes, and whether the connector was published to the user or group.

Clean up resources

To avoid future charges or unused access paths, remove only the resources that you created or changed for this walkthrough. These steps reverse the Amazon Quick knowledge base, Slack action connector, S3 access point, and IAM policy changes from the preceding sections.

1. Delete the Weekly Business Reporting Assistant skill from the Amazon Quick desktop application if you no longer need it.
2. Delete the Business Reporting Archive knowledge base if it was created only for this post.
3. Remove the Amazon S3 integration from Amazon Quick if no other knowledge bases use it.
4. Delete the S3 access point if it was created only for this workflow and no other workload uses it.
5. Remove IAM policy statements and access point policies that were added only for this walkthrough.
6. Remove sample files from the approved folder path if they were used only for testing.

Conclusion

You now have a governed weekly reporting workflow that creates cited business artifacts from file content stored on Amazon FSx for NetApp ONTAP. The S3 access point exposes only the approved folder path, and Amazon Quick indexes that content through the Amazon S3 integration. Business users can research reporting questions, generate reports and visuals, and prepare Slack summaries from one workspace without moving the source files out of the managed storage workflow.

For executives and business stakeholders, this pattern improves trust in recurring updates. Users review citations before they use the output and approve Slack posts before distribution. For builders, it keeps the first implementation narrow enough to validate governance, access, and reporting quality.

Start with one approved folder, one reporting use case, and one reviewed distribution channel. After users trust the citations and output quality, expand the knowledge base to more business units, add document-level ACLs if required, and refine the skill instructions for your internal reporting standards.

For more information

To get started, identify one approved reporting folder on your FSx for ONTAP volume and work with your storage team to create an S3 access point for that path. Then create the Amazon Quick knowledge base and reporting assistant skill, and validate the workflow with your first weekly update.

For more information, refer to the following resources:

---

About the authors

Ebbey Thomas

Ebbey is a Senior Generative AI Specialist Solutions Architect at AWS. He designs and implements generative AI solutions that address specific customer business problems. He is recognized for simplifying complexity and delivering measurable business outcomes for clients. Ebbey holds a BS in Computer Engineering and an MS in Information Systems from Syracuse University.

Eugene Thomas

Eugene is a Technical Account Manager at AWS focused on agentic AI, no-code automation, resilience, security, and cost optimization. With more than 10 years in customer-facing roles, he helps builders and business stakeholders turn complex cloud topics into practical solutions. He is also an active member of the Amazon Quick community, exploring how chat agents can simplify collaboration.