🏷️🤖 Vagas Labeler Bot

Automatic label classification for developer job postings — reads issues opened on backend-pt/vagas, dotnetdevbr/vagas, frontend-pt/vagas, and nodejsdevbr/vagas, and applies the labels each repository already has. Never creates, renames, or recolors a label — only applies and removes.

Full design rationale, the per-repo configuration schema, and the classification tiers are documented on the docs site.

GitHub issues ──HTTPS──► Webhook Ingress ──publish──► LavinMQ ──► Worker ──► GitHub REST API
                              │                                      │
                              └──────────── persist raw ─────────────┴──► MariaDB ◄── Private API

Three independently deployable entry points, Clean Architecture underneath:

src/
├── ApiBr.Vagas.Domain            # entities, value objects — zero dependencies
├── ApiBr.Vagas.Application       # use cases, ports (interfaces)
├── ApiBr.Vagas.Contracts         # AMQP message envelopes
├── ApiBr.Vagas.Configuration     # per-repo YAML schema, parser, validator
├── ApiBr.Vagas.Labeling          # deterministic classification engine
├── ApiBr.Vagas.Inference         # optional LLM tier — off by default
├── ApiBr.Vagas.Infrastructure    # MariaDB, LavinMQ, GitHub client, health/version
├── ApiBr.Vagas.Webhook           # entry point 1 — ingress (ASP.NET Core minimal API)
├── ApiBr.Vagas.Worker            # entry point 2 — AMQP consumer + scheduled jobs
└── ApiBr.Vagas.Api               # entry point 3 — private API (internal only)

Dependency direction: Webhook/Worker/Api → Infrastructure → Application → Domain. Domain never references anything else.

git clone https://github.com/ApiBR/vagas-labeler-bot.git
cd vagas-labeler-bot
dotnet build
dotnet test

Requires the .NET 10 SDK (see global.json for the pinned version). Running the entry points locally against real MariaDB/LavinMQ instances is covered in Getting Started on the docs site.

This repository is being built in phases; see the roadmap for the full plan and current status. In short:

• ✅ Skeleton — solution, Clean Architecture project graph, GitVersion, CI, health/version endpoints on all three entry points
• 🚧 Ingress, broker, persistence, config subsystem, classification engine, inference tier, private API, deployment — not yet built

1. 🍴 Fork the repository
2. 🌿 Create a feature branch: git checkout -b feature/my-feature
3. ✅ Commit your changes and ensure dotnet build / dotnet test pass
4. 📬 Push and open a pull request against main

All PRs run through CI (build + test + coverage) and SonarCloud analysis.

MIT © API BR