# AI use cases for Vibe Coding Toolkit (soumatheusgomes/vibe-coding-toolkit) Repo: [soumatheusgomes/vibe-coding-toolkit](https://github.com/soumatheusgomes/vibe-coding-toolkit) | Use cases: 4 | Page: https://dedupedai.com/use-cases/soumatheusgomes-vibe-coding-toolkit/ A collection of documentation, prompts, and templates that set up a machine or a project for building with AI coding agents such as Claude Code and Codex. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Start a Next.js project from a product idea: https://dedupedai.com/use-cases/soumatheusgomes-vibe-coding-toolkit/start-nextjs-project-from-idea/ - Bring an existing project up to the toolkit setup: https://dedupedai.com/use-cases/soumatheusgomes-vibe-coding-toolkit/upgrade-existing-project-with-toolkit/ - Prepare a development machine for agent-driven work: https://dedupedai.com/use-cases/soumatheusgomes-vibe-coding-toolkit/prepare-development-machine/ - Delegate a feature to an agent with checks at each stage: https://dedupedai.com/use-cases/soumatheusgomes-vibe-coding-toolkit/run-scoped-agent-task-with-workflow/ ## How to start a Next.js project from a product idea with Vibe Coding Toolkit Repo: [soumatheusgomes/vibe-coding-toolkit](https://github.com/soumatheusgomes/vibe-coding-toolkit) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README Turn a short product description into a configured Next.js foundation with product docs and checks. **Who it is for:** A founder or developer with a new product idea who wants a coherent starting point before writing features. **The problem:** Starting a new app means making many setup choices (framework, tests, linting, docs, agent instructions) before any product work happens. **The outcome:** A Next.js project with PRODUCT.md, the official foundation, documentation, agent instructions, and validation, with the next MVP task identified. ### Steps 1. Create or open an empty directory for the project and start Claude Code or Codex there. 2. Paste the two-line prompt from the README that points to docs/prompts/11-next-project-bootstrap.md. 3. Replace the second line with your project name, problem, audience, main flows, and constraints. 4. Let the agent run the setup and review the generated PRODUCT.md and docs before continuing. 5. Confirm the agent reports the checks it ran and the next MVP task. ### Prompt to paste into your agent ```text Read the repo soumatheusgomes/vibe-coding-toolkit at docs/prompts/11-next-project-bootstrap.md and execute the setup in this directory for the product described as . Report the generated files, the checks that passed, and the next MVP task. Verify that PRODUCT.md exists and that the validation step ran. ``` ### Caveat The README says the shortcut works from the published main branch, so in an unpublished checkout you must use the local prompt file instead. Risk, remote script: The agent fetches the prompt from a remote URL and follows its instructions, so review the fetched file before letting it run commands. 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/soumatheusgomes-vibe-coding-toolkit/start-nextjs-project-from-idea/ ## How to bring an existing project up to the toolkit setup with Vibe Coding Toolkit Repo: [soumatheusgomes/vibe-coding-toolkit](https://github.com/soumatheusgomes/vibe-coding-toolkit) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README Diagnose a current codebase and add applicable agent configuration without breaking its stack. **Who it is for:** A developer maintaining a project that already has code, dependencies, and some agent instructions, who wants consistent setup and checks. **The problem:** Existing projects often have partial or outdated agent configuration, and manual adoption risks duplicating instructions or breaking working contracts. **The outcome:** A diagnosis, a short list of improvements with completion criteria, applied configuration, and a before and after verification record with pending work noted. ### Steps 1. Open Claude Code or Codex at the application root. 2. Paste the prompt from the README that points to docs/prompts/12-existing-project-setup.md, asking it to execute the entire Prompt section starting with brainstorming. 3. Review the proposed improvements, risks, and completion criteria before approving changes. 4. Check the verification report for tested integrations and recorded limits. 5. Rerun the same prompt later as routine maintenance. ### Prompt to paste into your agent ```text Read the repo soumatheusgomes/vibe-coding-toolkit at docs/prompts/12-existing-project-setup.md and execute the Prompt section in the project at , starting with brainstorming. Report the before and after state, the changes applied, and any pending work. Verify that existing stack, database, and contracts were preserved. ``` ### Caveat The README says incompatible migrations need their own decision, so this prompt will not change the database or stack on its own. Risk, remote script: The prompt is fetched from a remote URL and drives changes across the project, so review the proposed changes before approving them. 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/soumatheusgomes-vibe-coding-toolkit/upgrade-existing-project-with-toolkit/ ## How to prepare a development machine for agent-driven work with Vibe Coding Toolkit Repo: [soumatheusgomes/vibe-coding-toolkit](https://github.com/soumatheusgomes/vibe-coding-toolkit) | Difficulty: Intermediate | Time: 90 minutes | Basis: From the README Install and record the tools, agents, and integrations needed before starting projects. **Who it is for:** A developer setting up a new computer or refreshing an older one for AI-assisted coding. **The problem:** Setting up system tools, Node.js, Docker, agents, plugins, and integrations by hand is slow and hard to repeat or audit later. **The outcome:** A machine with the runtime and agent tools in place, plus a record of versions, scopes, backups, and validations. ### Steps 1. Open an agent with terminal access on the computer you want to prepare. 2. Paste the prompt from the README that points to docs/prompts/10-development-environment-setup.md. 3. Let the agent run the entire Prompt section block. 4. Note any interventions the agent says it cannot complete itself. 5. Keep the generated manifest so later project setups can reuse the environment. ### Prompt to paste into your agent ```text Read the repo soumatheusgomes/vibe-coding-toolkit at docs/prompts/10-development-environment-setup.md and execute the entire Prompt section on this computer. Report installed versions, backups made, validations passed, and any required manual steps. Verify that the manifest of versions and scopes was written. ``` ### Caveat The README gives no details on operating systems or privileges needed, so confirm the agent can complete installs on your computer before starting. Risk, remote script: The agent installs system tools and plugins based on instructions fetched from a remote URL, so check what it plans to install first. 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/soumatheusgomes-vibe-coding-toolkit/prepare-development-machine/ ## How to delegate a feature to an agent with checks at each stage with Vibe Coding Toolkit Repo: [soumatheusgomes/vibe-coding-toolkit](https://github.com/soumatheusgomes/vibe-coding-toolkit) | Difficulty: Advanced | Time: 90 minutes | Basis: From the README Use the toolkit's workflow to move a request from design to verified, reviewed code. **Who it is for:** A developer who wants an agent to build a bounded feature in an existing project and needs clear checkpoints before accepting the change. **The problem:** Agents often start coding before the design is settled, or claim success without tests, which makes the results hard to trust. **The outcome:** A change that went through understand, design, implement, verify, and review stages, with evidence rather than a claim that it looks ready. ### Steps 1. Describe the feature and its boundaries to the agent in the project directory. 2. Ask the agent to brainstorm and propose a design before it edits files. 3. Approve a plan that names permitted files and acceptance criteria. 4. Require the agent to run the project checks and report results, not just the outcome. 5. Review the diff and the verification notes before merging. ### Prompt to paste into your agent ```text In the repo soumatheusgomes/vibe-coding-toolkit workflow, implement the feature in this project at . Follow understand, design, plan, implement, verify, and review in order, and keep changes to the smallest complete change. Report the diff summary and the checks run. Verify that the checks passed and that pending work is listed. ``` ### Caveat The README describes these tools as working agreements, so the quality gain depends on the agent actually running the checks and the project having tests to run. 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/soumatheusgomes-vibe-coding-toolkit/run-scoped-agent-task-with-workflow/