agent-memory
3 articles
Agent Memory Is Not Just Better RAG: What Grep and AKBP Are Really Saying
An arXiv paper found that inline grep often beats vector retrieval on long-memory conversational QA, while AKBP turns agent memory into a local-first, review-gated, file-backed protocol. Together, they point to the same lesson: agent memory is not a search feature. It is systems engineering.
Autobrowse: What Browser Agents Really Lack Is Not Brains, but Handoff-Ready Memory
Kyle Jeong introduces Browserbase's internal Autobrowse: browser agents repeatedly execute tasks on real websites, study their own traces, and graduate successful paths into readable, auditable, reusable skills.
One `message Romain` prompt runs the whole workflow — OpenAI DevX demos Codex Chronicle, but the costs the tweet skipped matter too
OpenAI DevX's Dominik Kundel says: now that Codex has memories, plugins, and the newly-dropped Chronicle, he no longer packages context for AI — one line 'sync docs + message Romain' reads a Google Doc, edits markdown, opens a PR, and DMs the right person on Slack. Very nice. But the three costs written into official Chronicle docs were not in the tweet: macOS screen-recording permission, memories stored unencrypted on device, prompt injection risk amplified. Chronicle is a screen-recording agent, not a harmless booster.