Resources

Primary sources first. Vendor documentation changes often; the links below point at the stable entry pages rather than specific sections.

Tool documentation

• Claude Code - Anthropic. Installation, CLAUDE.md, MCP, hooks, sub-agents.
• GitHub Copilot - GitHub. Repository instructions, agent mode, coding agent, code review, enterprise policies.
• Cursor - Cursor. Rules, agent mode, background agents, MCP, Bugbot.
• Model Context Protocol - the specification and the registry of servers.

Specification and testing

• Gojko Adzic, Specification by Example (Manning, 2011). The book behind Given / When / Then and example-driven specs; everything in Module 2 is downstream of it.
• Kent Beck, Test-Driven Development: By Example (Addison-Wesley, 2002). The original test-first contract.
• Hypothesis (Python) and fast-check (JavaScript / TypeScript) - property-based testing libraries.

RAG and evaluation

• pgvector - vector similarity search in PostgreSQL.
• Lewis et al., "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks" (2020) - the paper that named the pattern.
• Liu et al., "Lost in the Middle: How Language Models Use Long Contexts" (2023) - why "just include everything" fails.
• Ragas and promptfoo - open-source evaluation frameworks for LLM-backed features.

Security and governance

• OWASP Top 10 for Large Language Model Applications - prompt injection is LLM01.
• Simon Willison's writing on prompt injection - the clearest ongoing explanation of why it is a design problem.
• NIST AI Risk Management Framework - vocabulary that auditors recognise.
• EU Artificial Intelligence Act - the text and a navigable summary.
• GitHub dependency review action - licence and vulnerability gate for pull requests.

Measurement and adoption

• DORA - the research programme and the four metrics.
• Forsgren, Humble, Kim, Accelerate (IT Revolution, 2018). The book-length argument for measuring delivery outcomes.
• SPACE framework (Forsgren et al., 2021) - why individual productivity metrics fail and what to use instead.

Teaching

• Richard Mayer, Multimedia Learning (Cambridge, 3rd ed. 2020). The evidence base for "one worked example, fewer than six boxes".
• Greg Wilson, Teaching Tech Together (free online). Practical pedagogy for technical instructors.