# AI use cases for Claude Quickstarts (anthropics/claude-quickstarts) Repo: [anthropics/claude-quickstarts](https://github.com/anthropics/claude-quickstarts) | Use cases: 4 | Page: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/ A collection of starter projects that show developers how to build deployable applications on the Claude API. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run a Claude customer support agent over a knowledge base: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/build-customer-support-agent-prototype/ - Analyze financial data by chat with interactive charts: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/analyze-financial-data-with-chat-charts/ - Post a weekday brief of Slack and GitHub activity: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/schedule-daily-slack-and-github-brief/ - Automate browser tasks with Claude and Playwright: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/automate-browser-tasks-with-claude/ ## How to run a Claude customer support agent over a knowledge base with Claude Quickstarts Repo: [anthropics/claude-quickstarts](https://github.com/anthropics/claude-quickstarts) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Get an AI support agent that answers customer questions from a knowledge base, as a starting point to customize. **Who it is for:** A developer or support lead at a company that wants an AI-assisted support system and needs a working base to adapt. **The problem:** Building a support agent from scratch means wiring up Claude, a knowledge base and a chat interface before any answers can be tested. **The outcome:** A running customer support agent that answers from a knowledge base, which the team can then adapt to its own content and policies. ### Steps 1. Clone the Claude Quickstarts repository to your machine. 2. Open the customer-support-agent directory. 3. Install the dependencies described in that project's README. 4. Set your Claude API key as an environment variable. 5. Run the quickstart application and ask it a sample customer question. ### Prompt to paste into your agent ```text Work in the anthropics/claude-quickstarts repository. Open the customer-support-agent quickstart, follow its README to install dependencies and set the Claude API key, then start the app. Report the URL or command used to run it and the answer to one sample customer question, and confirm the answer draws on the knowledge base rather than generic text. ``` ### Caveat The README gives only a general overview of this project, so check its own README for knowledge base setup and any further configuration. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/build-customer-support-agent-prototype/ ## How to analyze financial data by chat with interactive charts with Claude Quickstarts Repo: [anthropics/claude-quickstarts](https://github.com/anthropics/claude-quickstarts) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Ask questions about financial data in chat and see interactive visualizations as answers. **Who it is for:** An analyst or finance-minded developer who needs to explore financial data quickly without writing analysis code first. **The problem:** Exploring financial data usually means writing scripts or building dashboards before the first useful chart appears. **The outcome:** A chat interface where questions about financial data return analysis with interactive data visualization. ### Steps 1. Clone the Claude Quickstarts repository to your machine. 2. Open the financial-data-analyst directory. 3. Install the dependencies described in that project's README. 4. Set your Claude API key as an environment variable. 5. Run the app and ask a question about your financial data in chat. ### Prompt to paste into your agent ```text Work in the anthropics/claude-quickstarts repository. Open the financial-data-analyst quickstart, follow its README to install dependencies, set the Claude API key and start the app. Ask it one question about the sample financial data and report whether it returned an analysis with an interactive chart. ``` ### Caveat The README does not say which data sources are supported, so confirm the input format in the project's own README before using real financial data. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/analyze-financial-data-with-chat-charts/ ## How to post a weekday brief of Slack and GitHub activity with Claude Quickstarts Repo: [anthropics/claude-quickstarts](https://github.com/anthropics/claude-quickstarts) | Difficulty: Advanced | Time: 2 hours | Basis: From the README Deploy a scheduled agent that posts one short brief to Slack each weekday from channels and pull requests. **Who it is for:** A team lead or engineer who wants a daily digest of Slack channel activity and open GitHub pull requests without checking each source by hand. **The problem:** Keeping up with several Slack channels and GitHub pull requests every morning takes time, and failed or missing sources are easy to overlook. **The outcome:** One short brief posted to Slack each weekday, with any source that could not be read reported as unreadable rather than silently skipped. ### Steps 1. Clone the Claude Quickstarts repository to your machine. 2. Open the managed-agents/daily-brief directory and read its README. 3. Set up the Slack and GitHub access, and the Claude API key, that the README requires. 4. Apply the deployment files the README describes, which are defined as files applied with ant apply. 5. Check the first scheduled post in Slack and confirm it lists each source. ### Prompt to paste into your agent ```text Work in the anthropics/claude-quickstarts repository. Open the managed-agents/daily-brief quickstart, follow its README to configure Slack, GitHub and the Claude API key, and apply the deployment files it describes. Report the deployment status and the first brief posted to Slack, and verify that every source is listed as either read or unreadable. ``` ### Caveat The README names the deployment files and the ant apply step but not the exact commands or permissions, so read the project README before applying anything to your workspace. Risk, credentials: The agent needs Slack and GitHub access, so store those tokens in the vault the deployment uses and never in files committed to the repository. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/schedule-daily-slack-and-github-brief/ ## How to automate browser tasks with Claude and Playwright with Claude Quickstarts Repo: [anthropics/claude-quickstarts](https://github.com/anthropics/claude-quickstarts) | Difficulty: Intermediate | Time: 1 hour 30 minutes | Basis: From the README Let Claude navigate websites, read page content and fill forms through a Playwright-backed browser tool. **Who it is for:** A developer who needs to automate repetitive web tasks such as reading pages or filling forms and wants a reference implementation to adapt. **The problem:** Writing browser automation for each site by hand is slow, and it is hard to tell how an AI model should drive a browser safely. **The outcome:** A working browser automation demo in which Claude navigates a site, inspects elements, extracts content and fills a form. ### Steps 1. Clone the Claude Quickstarts repository to your machine. 2. Open the browser-use-demo directory and read its README. 3. Install the dependencies described in that README. 4. Set your Claude API key as an environment variable. 5. Run the demo and give Claude a simple task, such as extracting content from one page. ### Prompt to paste into your agent ```text Work in the anthropics/claude-quickstarts repository. Open the browser-use-demo quickstart, follow its README to install dependencies, set the Claude API key and start the demo. Give it one read-only task on a page you control and report whether it extracted the expected content. ``` ### Caveat The README describes this as a reference implementation, so review the project's own guidance before pointing it at production sites or logged-in accounts. Risk, reverse engineering: Automated browsing can reach sites and accounts you do not own, so run it only against pages you are allowed to automate and review what it submits in forms. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/anthropics-claude-quickstarts/automate-browser-tasks-with-claude/