GH-600 GitHub Agentic AI Developer - Hub de Conteúdo
CertyPro/certy-gh600-course-content
Curso prático de agentes IA autônomos governados via GitHub
Ver no GitHubHub de ensino completo para o curso GH-600 que abrange construção, supervisão e governança de agentes autônomos em fluxos de entrega via GitHub. Inclui mapa de lições, objetivos de exame, walkthroughs de laboratório, cheat sheets, práticas e projeto capstone cobrindo arquitetura de agentes, MCP servers, memória, orquestração multi-agente e conformidade.
Instalação
npx convez-ai CertyPro/certy-gh600-course-content1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Aprender arquitetura e orquestração de agentes IA autônomos
- Dominar Model Context Protocol (MCP) servers e tooling
- Implementar memória, estado e controle via GitHub Issues/Actions
- Preparar-se para o exame oficial GH-600 com mock tests e capstone
Veredito

Recurso educacional robusto e estruturado para desenvolvimento de agentes IA em produção. Material bem organizado com progressão clara (teoria → labs hands-on → validação), mas requer experiência prévia em desenvolvimento e arquitetura de sistemas para máximo aproveitamento.
README
A tradução para o português deste README ainda está na fila.
GH-600 GitHub Agentic AI Developer - Course Content
Build, supervise, and govern autonomous agents in GitHub-driven delivery.
This is the public teaching hub for the GH-600 GitHub Agentic AI Developer course by Certy, built and maintained by Coded Vision Design. It contains the lesson map, exam objectives, instructor lab walkthroughs, cheat sheets, practice questions, a glossary, diagrams and the capstone brief.
The free written course and the mock exam live at certy.pro. The hands-on practice repositories live under the CertyPro organisation.
The repo set
The course is delivered across one content hub (this repo) and seven hands-on lab repositories. Each lab maps to an exam domain.
| Repo | Purpose | Domain |
|---|---|---|
| certy-gh600-course-content | This hub - objectives, lesson map, cheat sheets, labs, glossary, capstone | All |
| certy-gh600-agent-starter | Minimal agent scaffold (planner, executor, validator) you extend throughout the course | 1.0 Agent Architecture, Planning & Validation |
| certy-gh600-mcp-server-lab | Build and connect Model Context Protocol servers, tools and resources over stdio and SSE | 2.0 Tooling, MCP Servers & Execution Context |
| certy-gh600-memory-state-lab | Short-term, episodic and semantic memory; state serialisation and checkpoints | 3.0 Memory, State & GitHub Context |
| certy-gh600-github-control-plane-lab | Use GitHub issues, PRs, Actions and the API as the agent's control plane and context source | 3.0 Memory, State & GitHub Context |
| certy-gh600-evaluation-debugging-lab | Eval datasets, rubrics, trace inspection and the iteration loop | 4.0 Evaluation, Debugging & Iteration |
| certy-gh600-multi-agent-orchestration-lab | Orchestrator and worker agents, handoffs, state locking and conflict resolution | 5.0 Multi-Agent Systems & Orchestration |
| certy-gh600-governance-guardrails-lab | Input safeguards, output redaction, approval gates and audit logs | 6.0 Governance, Guardrails & Compliance |
How to use this course
- Read the free course at certy.pro. Work through the 15 lessons in order; they map to the six exam domains.
- Clone the matching lab repo for each domain and complete the exercises. Use the instructor walkthroughs in
labs/as a guide. - Keep the cheat sheets open while you work - they distil the exam-critical points per domain.
- Self-test with the practice questions here, then sit the full mock exam at certy.pro.
- Finish with the capstone - extend the agent-starter into a supervised, governed agent that touches every domain.
- Book the official GH-600 exam when your mock-exam scores are consistently above the pass mark.
Contents
README.md- this overviewexam-objectives.md- the six domains, weights, skills and matching lessons and labslesson-map.md- the 15 Certy lessons mapped to chapters and reposvideo-roadmap.md- planned video walkthroughs, one per domainlabs/- instructor walkthrough notes for the six chapterscheat-sheets/- one revision sheet per domain topicpractice-questions/- sample MCQs and a pointer to the mock examglossary/- key agentic-AI, MCP and GitHub termsdiagrams/- ASCII diagrams of the agent loop and a multi-agent pipelineresources/- official cert, MCP and GitHub links plus all sibling reposchallenge-project/- the GH-600 capstone briefexam-objectives.md,CODE_OF_CONDUCT.md,LICENSE,LICENSE-CONTENT.md
Exam at a glance
| Domain | Weight |
|---|---|
| 1.0 Agent Architecture, Planning & Validation | 15% |
| 2.0 Tooling, MCP Servers & Execution Context | 25% |
| 3.0 Memory, State & GitHub Context | 25% |
| 4.0 Evaluation, Debugging & Iteration | 15% |
| 5.0 Multi-Agent Systems & Orchestration | 10% |
| 6.0 Governance, Guardrails & Compliance | 10% |
See exam-objectives.md for the full skill breakdown per domain.
Licensing
- Code in this repository and the sibling labs is licensed under the MIT licence.
- Prose (lessons, walkthroughs, cheat sheets, diagrams and questions) is licensed under CC-BY-4.0.
Attribution: Certy / Coded Vision Design. See LICENSE-CONTENT.md for details.
Maintained by Coded Vision Design for Certy. Contributions welcome - please read the Code of Conduct first.