Context engineering for coding agents involves curating what an LLM sees to improve results. The main categories include reusable prompts (instructions and guidance), context interfaces (tools, MCP servers, skills), and workspace files. Key considerations are who decides when to load context (LLM, human, or agent software),

8m read time From martinfowler.com
Post cover image
Table of contents
What is context in coding agents?If and when: Who decides to load context?How much: Keeping the context as small as possibleExample: Claude CodeSharing context configurationsBeware: Illusion of control

Sort: