This repository provides a reusable, customizable configuration template for agentic coding assistants such as:
- Cascade (Windsurf)
- Claude
- GitHub Copilot Workspaces
- Cline
- Custom local or server-based agents
It defines how an assistant should behave, what tools it can use, and how to interact with your codebase — especially for test- and automation-heavy projects.
- Playwright-first testing architecture
- Yarn-only package management (no npm)
- Page Object Model (POM) structure assumed
- Designed for macOS + ZSH
- Agent-agnostic: usable with any LLM coding assistant
.agentic/
├── README.md # Instructions for using and renaming the folder
├── system_prompt.md # Defines assistant behavior and expectations
├── tools.json # Describes tool interfaces the assistant can invoke
├── environment.json # Optional: environment hints like cwd, OS, shell
- Use
system_prompt.mdfor agent behavior - Parse and expose
tools.jsonfor controlled tool usage - Respect
environment.jsonif your runtime supports agent hints
| File | When to Update |
|---|---|
system_prompt.md |
When your test setup, workflows, or conventions change |
tools.json |
When you add/remove tool capabilities |
environment.json |
When your shell, OS, working directory, or package manager changes |
- Tests follow the Page Object Model structure
- Project uses Playwright for end-to-end automation
- Yarn is the only supported package manager
- Development occurs on macOS using the ZSH shell
- New scripts must:
- Be added to
package.jsonunderscripts - Be documented in the project’s root
README.md
- Be added to
This configuration is public-domain or MIT licensed (your choice).
Fork, clone, copy, and adapt it into any personal or commercial project.