# AI use cases for Alphacode (dragonked2/alphacode) Repo: [dragonked2/alphacode](https://github.com/dragonked2/alphacode) | Use cases: 5 | Page: https://dedupedai.com/use-cases/dragonked2-alphacode/ AlphaCode is a free, MIT-licensed AI coding agent written in Rust that runs in the terminal and can edit code, run commands and tests, drive Firefox, and control desktop apps. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Fix a failing test or bug in a repo from the terminal: https://dedupedai.com/use-cases/dragonked2-alphacode/fix-bug-with-terminal-agent/ - Test a web app flow in a live Firefox session: https://dedupedai.com/use-cases/dragonked2-alphacode/test-web-app-in-firefox/ - Drive a native desktop app with a written instruction: https://dedupedai.com/use-cases/dragonked2-alphacode/automate-desktop-app-steps/ - Run parallel analysis of a codebase with swarm mode: https://dedupedai.com/use-cases/dragonked2-alphacode/split-review-with-swarm/ - Trace untrusted input in a system you are authorized to test: https://dedupedai.com/use-cases/dragonked2-alphacode/review-owned-app-for-security-issues/ ## How to fix a failing test or bug in a repo from the terminal with Alphacode Repo: [dragonked2/alphacode](https://github.com/dragonked2/alphacode) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Give AlphaCode a bug report and let it inspect the code, edit files, and run tests. **Who it is for:** A developer whose test suite is failing or whose app has a regression and who wants a first-pass fix to review. **The problem:** Tracing a failure through unfamiliar code and running the right tests takes time away from the actual fix. **The outcome:** A proposed change to the repo, with the relevant tests run and a report of what changed and what was checked. ### Steps 1. Run `curl -fsSL https://raw.githubusercontent.com/dragonked2/alphacode/main/scripts/install.sh | bash` on macOS or Linux, or `irm https://raw.githubusercontent.com/dragonked2/alphacode/main/scripts/install.ps1 | iex` in PowerShell on Windows. 2. Run `alphacode doctor` to check your environment. 3. Change into your project folder and run `alphacode run "fix the failing test"`, or replace the quoted text with your own request. 4. Review the diff and the test results the agent reports before you commit anything. 5. Run `alphacode` to open the interactive UI if you want to follow up on the same task. ### Prompt to paste into your agent ```text Use the dragonked2/alphacode repo to fix the failing test in . Run the relevant tests after the change and report the changed files and test results. Verify that the failing test now passes and that no other tests were newly broken. ``` ### Caveat The README says the built-in free AI lane works without an API key, but connected providers may require their own credentials and may incur charges. 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/dragonked2-alphacode/fix-bug-with-terminal-agent/ ## How to test a web app flow in a live Firefox session with Alphacode Repo: [dragonked2/alphacode](https://github.com/dragonked2/alphacode) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Have AlphaCode walk through a login or form flow in Firefox and report reproducible problems. **Who it is for:** A QA tester or web developer who needs to check a JavaScript-heavy page or an authenticated flow that scripted tests do not cover well. **The problem:** Manual checks of dynamic pages and logged-in flows are slow and easy to skip, and simple HTTP tools cannot see what the page renders. **The outcome:** A walkthrough of the flow in a real browser, with screenshots and a list of reproducible problems. ### Steps 1. Install the AlphaCode Browser Agent extension from Firefox Add-ons at https://addons.mozilla.org/en-US/firefox/addon/alphacode-browser-agent/. 2. Run `alphacode browser setup`. 3. Run `alphacode browser status` and confirm the browser bridge responds. 4. Run `alphacode` and describe the flow to test, such as logging in, opening the dashboard, and checking form validation. 5. Review the reported results and screenshots, and confirm any problem by repeating it yourself. ### Prompt to paste into your agent ```text Use the dragonked2/alphacode repo with its Firefox browser automation to test the login flow at . Report each step, capture a screenshot of any failure, and list only problems you could reproduce. Verify the report by repeating one failing step manually. ``` ### Caveat The README says browser automation acts inside a real browser session and that page content should be treated as untrusted input. Risk, credentials: Browser automation can act inside an existing authenticated session, so review the requested actions before letting the agent submit forms or change account settings. 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/dragonked2-alphacode/test-web-app-in-firefox/ ## How to drive a native desktop app with a written instruction with Alphacode Repo: [dragonked2/alphacode](https://github.com/dragonked2/alphacode) | Difficulty: Advanced | Time: 60 minutes | Basis: From the README Ask AlphaCode to open a native app, perform steps, and read a result through accessibility APIs. **Who it is for:** A user who repeats the same click-and-read steps in a native desktop application and wants them handled from a single instruction. **The problem:** Repetitive desktop tasks with no API are tedious and error-prone when done by hand. **The outcome:** The requested steps completed in the desktop app, with the result read back and reported. ### Steps 1. Run the install command from the Install section for your platform, then run `alphacode doctor`. 2. Confirm the accessibility backend for your OS is available: UI Automation on Windows, the Accessibility API on macOS, or AT-SPI2 on Linux. 3. Run `alphacode` and give the instruction, for example: Open Calculator, calculate 123 × 456, and read the result. 4. Check that the result matches the expected value of 56088 and that the agent did not change other windows. 5. Repeat the task with a different target app once the first run works. ### Prompt to paste into your agent ```text Use the dragonked2/alphacode repo's desktop automation to open , complete the steps: , and read the result. Report the value read from the screen. Verify the result by comparing it with a value you can calculate yourself. ``` ### Caveat The README says desktop interaction depends on a working accessibility backend and on the target application being available. 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/dragonked2-alphacode/automate-desktop-app-steps/ ## How to run parallel analysis of a codebase with swarm mode with Alphacode Repo: [dragonked2/alphacode](https://github.com/dragonked2/alphacode) | Difficulty: Advanced | Time: 60 minutes | Basis: From the README Split one review goal into separate analysis tasks and collect the combined results. **Who it is for:** A tech lead preparing a broad review of an application before a release, who wants architecture, security, testing, and documentation covered at once. **The problem:** A single agent working through a large review in sequence is slow, and parallel reviewers often edit the same files. **The outcome:** A combined review report covering several perspectives, produced by coordinated agents. ### Steps 1. Install AlphaCode and run `alphacode doctor`. 2. Change into the repository root you want reviewed. 3. Run `alphacode` and enter `/swarm "Analyze this application from architecture, security, testing, and performance perspectives."`, adjusting the perspectives to your needs. 4. Read the coordinator's combined report and check which findings reference real files. 5. Ask for follow-up fixes in a separate run so reviewers do not change the same files at once. ### Prompt to paste into your agent ```text Use the dragonked2/alphacode repo's swarm mode on to review architecture, testing, and performance in parallel. Return one combined report with file references for each finding. Verify that at least three findings point to real lines in the repository. ``` ### Caveat The README notes swarm mode works best when tasks can be divided so agents do not modify the same files at the same time. 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/dragonked2-alphacode/split-review-with-swarm/ ## How to trace untrusted input in a system you are authorized to test with Alphacode Repo: [dragonked2/alphacode](https://github.com/dragonked2/alphacode) | Difficulty: Advanced | Time: 60 minutes | Basis: From the README Have AlphaCode trace input to sensitive operations and confirm which findings reproduce. **Who it is for:** A security engineer or bug bounty tester with written authorization to assess a specific application. **The problem:** Following untrusted input through a codebase by hand is slow, and unverified findings waste triage time. **The outcome:** A list of likely issues, each marked as reproduced or not reproduced in the authorized test environment. ### Steps 1. Confirm in writing that you are authorized to test the target system. 2. Install AlphaCode with the one-line installer and run `alphacode doctor`. 3. Run `alphacode` in the target repository and give the goal: trace untrusted input to sensitive operations and validate likely findings. 4. For any finding that needs a live check, use the Firefox browser setup with `alphacode browser setup` and `alphacode browser status`. 5. Keep only findings that were reproduced, and record the steps to reproduce each one. ### Prompt to paste into your agent ```text Use the dragonked2/alphacode repo to trace untrusted input to sensitive operations in . Report only findings that you reproduced in the authorized test environment, with the reproduction steps. Verify each reported issue by re-running its reproduction once. ``` ### Caveat The README limits security testing to systems you own or are explicitly authorized to assess, and the same workflow could be misused against systems you do not control. 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/dragonked2-alphacode/review-owned-app-for-security-issues/