Running Claude Code, Codex and Cursor on One Project: How I Finally Stopped Losing Context
Asly97 · reddit · 2026-09-18
A developer shares months of experience running Claude Code, Codex and Cursor on the same project — the tools got smart while memory stayed dumb.
His three-layer approach:
- claude.md / agents.md for permanent context: stack, conventions, mistakes not to repeat;
- At each session's end, have the agent write 10 lines on progress, decisions and blockers, then paste it back next time — 30 seconds of effort saving 20 minutes of recall daily;
- One shared context store via an MCP server (vilix ai) so all three tools read the same state, enabling "plan in one, build in another."
Honest caveat: models are lazy and often skip checking memory unless prompted with "check memory first." His conclusion: stop shopping for tools — amnesia was the real bottleneck all along.
More from coding & agent
- 'Jev' agent plays Pac-Man with Astra strategizing and millisecond execution — daniel_mac8 · 2026-09-18
- Awesome Code-as-X: curated papers unifying agent research via programs as representations — ccloy · 2026-09-18
- 8 AI personal agents, zero successful SMS — 10DLC is the blocker — iamstanty · 2026-09-18
- Zalando details its Kubernetes-based internal agent platform with kagent and kro — bibryam · 2026-09-18
- Modern coding: fantasize a crazy concept on Twitter, then hand the link to a coding agent — airesearch12 · 2026-09-18
- Dev swapped to Opus 5 after OpenAI credits ran out — 'this is not great' — lucasmeijer · 2026-09-18