Guides
Step-by-step walkthroughs for common jobs.
CLAUDE.md vs shared memory for coding agents
CLAUDE.md is instructions loaded into every session; shared memory is history an agent searches when it needs it. When to use each, with examples.
October 30, 2026Cursor rules for a team: project rules, team rules and shared memory
How a team shares Cursor rules: .cursor/rules in git, Team Rules from the dashboard, AGENTS.md, and a shared ledger for decisions the rules don't cover.
October 30, 2026How to give Claude Code memory between sessions
Claude Code starts every session with a fresh context. What CLAUDE.md and auto memory carry over, where they stop, and how to add a shared project ledger.
October 30, 2026How to share AI coding agent context across a team
Each teammate's agent starts from zero. How to share instructions, decisions, lessons and tasks across Claude Code, Cursor and Codex in one team.
October 30, 2026How to stop AI coding agents repeating reverted changes
Agents redo changes you already reverted because the reason lives in a PR or a chat. Record it as a lesson they recall before the next change.
October 30, 2026