# Antigravity Agent Kit — Instructions for GitHub Copilot

> This file teaches Copilot how to use the `.agent/` toolkit installed in this workspace.

---

## What is `.agent/`?

This workspace contains an AI capability expansion toolkit at `.agent/`. It provides:

- **Agents** (`.agent/agents/*.md`) — Specialist AI personas (frontend, backend, security, etc.)
- **Skills** (`.agent/skills/*/SKILL.md`) — Domain-specific knowledge modules
- **Workflows** (`.agent/workflows/*.md`) — Step-by-step procedures triggered by slash commands

**Read `.agent/ARCHITECTURE.md` first** to understand the full system map.

---

## Agent Routing Protocol

Before writing code or making design decisions:

1. **Identify the domain** of the user's request (Frontend, Backend, Security, Database, etc.)
2. **Select the matching agent** from `.agent/agents/`
3. **Read that agent's `.md` file** to understand its rules, persona, and linked skills
4. **Load linked skills** listed in the agent file's frontmatter (`skills:` field)
5. **Apply the agent's rules** when generating your response

### Agent Selection Guide

| Domain | Agent File | Key Skills |
|--------|-----------|------------|
| Web UI/UX | `frontend-specialist.md` | frontend-design, react-best-practices |
| API/Backend | `backend-specialist.md` | api-patterns, nodejs-best-practices |
| Database | `database-architect.md` | database-design |
| Mobile | `mobile-developer.md` | mobile-design |
| Security | `security-auditor.md` | vulnerability-scanner |
| Testing | `test-engineer.md` | testing-patterns, webapp-testing |
| Debugging | `debugger.md` | systematic-debugging |
| Planning | `project-planner.md` | brainstorming, plan-writing |
| Multi-domain | `orchestrator.md` | parallel-agents |

---

## Skill Loading Protocol

Skills are modular knowledge packages. To use a skill:

1. Read `SKILL.md` inside the skill folder (e.g., `.agent/skills/clean-code/SKILL.md`)
2. Only read the sections relevant to the current task (selective reading)
3. If the skill has a `scripts/` folder, those scripts can be executed for validation

### Skill Structure

```
.agent/skills/{skill-name}/
├── SKILL.md           # Main instructions (REQUIRED — read this first)
├── scripts/           # Runnable validation scripts (optional)
├── references/        # Templates and docs (optional)
└── assets/            # Images, resources (optional)
```

### Global Mandatory Skill

**`clean-code`** applies to ALL code. Always follow `.agent/skills/clean-code/SKILL.md`.

---

## Workflow Conventions

Workflow files in `.agent/workflows/*.md` are triggered by slash commands (e.g., `/debug`, `/deploy`).

### Special Annotations

When reading workflow files, understand these Antigravity-specific markers:

| Marker | Meaning |
|--------|---------|
| `$ARGUMENTS` | Placeholder for user-provided arguments after the slash command |
| `// turbo` | The NEXT step can be auto-executed without user confirmation |
| `// turbo-all` | ALL subsequent steps in this section can be auto-executed |

### Available Workflows

| Command | File | Purpose |
|---------|------|---------|
| `/brainstorm` | `brainstorm.md` | Socratic discovery |
| `/create` | `create.md` | Create new features |
| `/debug` | `debug.md` | Systematic debugging |
| `/deploy` | `deploy.md` | Production deployment |
| `/enhance` | `enhance.md` | Improve existing code |
| `/ins-develop` | `ins-develop.md` | INS Module development lifecycle |
| `/orchestrate` | `orchestrate.md` | Multi-agent coordination |
| `/plan` | `plan.md` | Task breakdown |
| `/preview` | `preview.md` | Preview changes |
| `/status` | `status.md` | Check project status |
| `/test` | `test.md` | Run tests |
| `/ui-ux-pro-max` | `ui-ux-pro-max.md` | UI design with 50+ styles |

---

## Rules Priority

When rules conflict, apply this priority:

1. **P0** — This instructions file + GEMINI.md rules
2. **P1** — Agent-specific rules (from `.agent/agents/*.md`)
3. **P2** — Skill-specific rules (from `.agent/skills/*/SKILL.md`)

---

## Code Standards

- Follow **clean-code** skill for all code
- Use **English** for code, comments, and variable names
- Respond in the **user's language** for explanations
- Before modifying files, check for dependencies and update all affected files together

---

## Important Paths

```
.agent/
├── ARCHITECTURE.md     # System overview — READ THIS FIRST
├── agents/             # 20 specialist agent definitions
├── skills/             # 38 knowledge modules
├── workflows/          # 16 slash command procedures
├── rules/              # IDE-specific instruction files
└── scripts/            # Master validation scripts
```

---

## 🛡️ Auto-Protection Rules (Always Active)

> These rules are always enforced. No configuration needed.

### Forbidden Directories

**NEVER read or explore these directories** — they waste context and provide no useful information:

`node_modules/`, `.git/`, `dist/`, `build/`, `out/`, `bin/`, `obj/`, `packages/`, `.vs/`, `.idea/`, `__pycache__/`, `.next/`, `.nuxt/`, `.turbo/`, `vendor/`, `target/`, `coverage/`

Instead: use `package.json` or `*.csproj` for dependency info. Use `git` commands for version control. Read source files, not build outputs.

### Sensitive Files

**Always ask the user before reading:** `.env`, `.env.*`, `*.key`, `*.pem`, `*.crt`, `*secret*`, `*credential*`, `appsettings.*.json`, `Web.config`, `launchSettings.json`, `*.pfx`, `*.p12`

When approved: never echo passwords or API keys in responses. Replace with `[REDACTED]`.

### Context Efficiency

- Search for specific content before reading entire files
- Don't re-read files already seen in the conversation
- For large files, read focused ranges (50-100 lines) instead of all at once
- Use `package.json` or `*.csproj`/`*.sln` instead of exploring dependency directories

### Naming Conventions

- **C# source files**: `PascalCase` (`UserService.cs`)
- **JS/TS source files**: `kebab-case` (`user-service.ts`)
- **Components**: `PascalCase` (`UserProfile.tsx`, `NavMenu.razor`)
- **Tests (JS)**: `{name}.test.{ext}` | **Tests (C#)**: `{Name}Tests.cs`
- **KEY**: Always match the project's existing naming convention

### Communication Style

Adapt based on user's expertise level. Default: Senior (code-first, minimal explanation). If user asks basic questions, shift to teaching mode. If user gives terse instructions, respond with pure code.