Resources¶
Keep learning after Building the AI Native Cloud at AgentCon Singapore 2026. Start with the learning paths supplied for this session, then use the reference documentation when implementing your own workflow.
Session learning paths¶
- Awesome Copilot Learning Hub: community-curated guidance and practical patterns.
- GitHub Copilot Fundamentals: build a foundation for responsible, effective Copilot use.
- GitHub Agentic Developer Learning Path: explore agentic development practices.
GitHub Copilot and the Agentic SDLC¶
- GitHub Copilot documentation: editor setup, features, plans, and administration.
- Repository custom instructions: provide shared, version-controlled development guidance.
- Copilot coding agent: understand delegated issue-to-pull-request workflows.
- Responsible use of GitHub Copilot: understand limitations and human review responsibilities.
- GitHub Actions documentation: automate tests, builds, and release gates.
Tools, evaluation, and secure systems¶
- Model Context Protocol: understand the protocol, then design your own permission boundaries.
- Python unittest: run and extend the workshop's deterministic test suite.
- OWASP Top 10 for LLM applications: explore prompt injection, excessive agency, and related risks.
- Azure Well-Architected Framework: connect AI workflows with reliability, security, and operational excellence.
Workshop source and website¶
- Workshop repository: Python example, tests, site configuration, and original slide images.
- MkDocs: build the website locally.
- Material for MkDocs: navigation, search, and accessible theme controls.
- GitHub Pages: understand the publishing workflow.
- AgentCon Singapore: conference information.
The workshop example is deliberately offline and read-only. Before using a real model or cloud API, add authenticated identity, scoped credentials, schema validation at the model boundary, deadlines, redacted observability, and a representative evaluation dataset.