Overview & result
Jetwani Avinash demonstrates a memory gate for Claude Code: Jev decides whether a message is worth remembering, and jevmem saves useful project memory in JEVMEM.md. Its public README describes retaining decisions, rules, bugs, to-dos and failed approaches, then bringing relevant lines into later sessions. The repository was inspected, not independently installed or tested.
From the original post
First thing I built on Jev: a memory gate for Claude Code.
Every message gets one question: is this worth remembering? 0.3s later it's either skipped or saved as one line in JEVMEM.md in the repo.
Open source, link in my pinned post.
How Jev fits in the loop
- A message reaches jevmem during the coding workflow.
- TypeSafe Jev decides whether it contains useful project memory.
- jevmem writes retained memory to JEVMEM.md and supplies relevant lines in later sessions, as described in the current README.
How to reproduce
- Open the original demonstration and the matching public jevmem repository.
- Follow the current README: a TypeSafe API key and a supported Claude Code version are required; install the CLI, save the key, enable it in a project, then run jevmem doctor.
- Review the saved JEVMEM.md lines and omissions using a small non-sensitive example before relying on the gate.
Why this build matters
Gives coding teams a concrete path for retaining decisions and avoiding repeated failed approaches.
Limitations
- The post’s 0.3-second timing and the README’s evaluation results are author-reported.
- Repository inspection is not a completed installation or an independent memory-quality benchmark.
- The current repository may contain features added after the original video.