memx: an open-source CLI that debugs agent memory failures via state snapshot diffs
vivek9patel · reddit · 2026-09-14
Developer vivek9patel released memx, an experimental open-source CLI for debugging agent memory engines (Mem0, Supermemory, Zep/Graphiti, custom vector/graph stores). Unlike benchmarks like LoCoMo that only grade final LLM responses, memx hooks into memory backends, takes pre/post-session state snapshots, and diffs them against expected updates to classify failures into four stages: failed extraction, ignored contradiction, failed invalidation of stale records, or retriever fetching the wrong node. Repo: github.com/vivek9patel/memx.
More from coding & agent
- Dev puts a 4,000-customer support inbox on autopilot with agents, stops reading code — natesiggard · 2026-09-14
- Economist slams Claude's coding judgment: skips existing work, denies mistakes, praises Astra — soumitrashukla9 · 2026-09-14
- Developer shares one-shot Replit prompt workflow that generates games with just a genre swap — amasad · 2026-09-14
- Hamel Husain: Codex can already do anything Muse-style agent tools offer — skip the tool sprawl — HamelHusain · 2026-09-14
- DeskRoot Rebuilds AI Assistant Setup as a Folder of Editable Markdown Procedures — KenGuy14 · 2026-09-14
- arscontexta launches early agent-native IDE for typed knowledge bases — blaizedsouza · 2026-09-14