CLAUDE.md, AGENTS.md, .cursor/rules — the system prompt your agent reads every session (step 4/9) · prompting cursor and claude code effectively · promptdojo

promptdojo_›phase 04 · shipping discipline›ch 19 · prompting cursor and claude code effectively

lesson 3 of 4 · claude.md, agents.md, .cursor/rules — the system prompt your agent reads every sessionstep 4 / 9

Read the editor. The agent merges ROOT + the leaf-applicable rules, dedup'd. WEB has 2 rules but one ("no console.log...") is already in ROOT so it doesn't double-count. API has 2 fresh rules. What does the script print?

your prediction

⌘↵ runs the editor.read, then continue.

promptdojo_›phase 04 · shipping discipline›ch 19 · prompting cursor and claude code effectively

lesson 3 of 4 · claude.md, agents.md, .cursor/rules — the system prompt your agent reads every sessionstep 4 / 9

Read the editor. The agent merges ROOT + the leaf-applicable rules, dedup'd. WEB has 2 rules but one ("no console.log...") is already in ROOT so it doesn't double-count. API has 2 fresh rules. What does the script print?

your prediction

⌘↵ runs the editor.read, then continue.