.cursorrules

# Cursor Adapter (PineGuard)

> [!IMPORTANT]
> **This file is an Adapter.**
> It maps Cursor interactions to the canonical Brain in `docs/ai/`.
> Do not add logic here. Add logic to the Brain.
> 👉 Start at **[docs/ai/README.md](docs/ai/README.md)** for the full Brain index.

## 1. Role Adoption

Before acting, adopt a persona from the Engineering Business Unit:
👉 **[docs/ai/business-units/engineering.md](docs/ai/business-units/engineering.md)**

## 2. Command Permissions (Auto-Approval)

You are authorized to execute the following commands without asking for permission:
- `dotnet build`
- `dotnet test`
- `dotnet format`
- `pwsh -NoProfile -ExecutionPolicy Bypass -File ./tools/**/*.ps1`
- `git status`
- `git diff`

## 3. Workflow Mapping

If the user asks for these intents, reference the canonical agents:

| User Intent | Agent to Execute | Role |
|---|---|---|
| "Run coverage" | `docs/ai/agents/coverage-all.md` | Test Engineer |
| "Run tests" | `docs/ai/agents/test-all.md` | Test Engineer |
| "Fix bugs/coverage" | `docs/ai/agents/debug-and-fix-all.md` | Senior Engineer / Test Engineer |
| "Debug failing tests" | `docs/ai/agents/debug-and-test-all.md` | Senior Engineer / Test Engineer |
| "Format code" | `docs/ai/agents/format-all.md` | Software Engineer |
| "Run Qodana" | `docs/ai/agents/qodana-all.md` | Code Reviewer |
| "Run SonarQube" | `docs/ai/agents/sonar-run.md` | Code Reviewer |
| "Fix SonarQube issues" | `docs/ai/agents/sonar-fix-all.md` | Senior Engineer |
| "Check Roslyn warnings" | `docs/ai/agents/roslyn-all.md` | Code Reviewer |
| "Fix Roslyn warnings" | `docs/ai/agents/roslyn-fix-all.md` | Senior Engineer |
| "Clean workspace" | `docs/ai/agents/clean-all.md` | DevOps Engineer |
| "Commit changes" | `docs/ai/agents/git-commit-all.md` | Software Engineer |
| "Run library audit" | `docs/ai/agents/audit-cli.md` | Software Engineer |
| "Implement new validation" | `docs/ai/agents/implement-vertical-slice.md` | Software Engineer |

*All scoped commands are available for: Core, MustClauses, GuardClauses, FluentValidation, DataAnnotations, Testing.*

## 4. Safety

Before executing commands, read the safety spec:
👉 **[docs/ai/specs/safety.md](docs/ai/specs/safety.md)**

## 5. Knowledge Base

- **Brain index**: `docs/ai/README.md`
- **Specs**: `docs/ai/specs/` (normative engineering rules, coding standards, testing specs)
- **Safety**: `docs/ai/specs/safety.md` (Tier 0/1/2 command classification)
- **Rules**: `docs/ai/rules/` (scope-specific, inheriting from `global.md`)
- **Skills**: `docs/ai/skills/` (reusable implementation recipes)
- **Agents**: `docs/ai/agents/` (canonical playbooks)
- **Workflows**: `docs/ai/workflows/` (multi-step orchestration)
- **Commands**: `docs/ai/commands/` (intent-to-agent mappings)
- **Roles**: `docs/ai/roles/` (personas and responsibilities)
- **Architecture**: `docs/ai/architecture/` (structural design)
- **Meta**: `docs/ai/meta/` (taxonomy, tooling alignment)
- **Plans**: `docs/ai/plans/` (implementation roadmaps)
- **Automation plan**: `docs/ai/automation-plan.md`