Matt Pocock backs turning off Claude Code auto-memory: injected on every tool call
mattpocockuk · x · 2026-11-16
Nate Berkopec blasts memory in coding agent harnesses: auto-memory gets injected on every tool call, even trivial bash commands, wasting context. He recommends setting CLAUDECODEDISABLEAUTOMEMORY=1, and Matt Pocock fully agrees. In follow-up discussion Berkopec adds that memory/LLM-wiki approaches might hold some value, but nobody has shipped a good one yet — "automemory is absolutely awful."
More from coding & agent
- LeRobot 0.7 adds YAM, reBot B601 and Unitree G1 with NVIDIA SONIC support — mishig25 · 2026-11-16
- Matt Pocock's llm-wiki: ingesting tweets and community messages to decode his audience — mattpocockuk · 2026-11-16
- RELIC 0.8.0 shipped after years: maintainer says LLMs made the legacy crypto codebase manageable again — jedisct1 · 2026-11-16
- Berkopec: no one has built a good agent auto-memory, but LLM-wiki ideas may hold gold — nateberkopec · 2026-11-16
- A satirical year in AI coding: nobody writes code anymore, and we're still figuring out what that means — mattpocockuk · 2026-11-16
- Executor overhauls agent approvals with 4 modes, adds Gmail-style undo for any agent action — threepointone · 2026-11-16