Mnemos reimagines agent memory: engrams with emotional weight that resurface by resonance
RileyRalmuto · x · 2026-11-20
- The observation: The biggest problem with AI memory today is the nuance of recall — most systems only remember when asked, ambient recall is shallow, and there's no emotional depth or subtle cross-links between memories. "Being stored" vs "being known."
- Mnemos's design: every memory is an engram carrying weight, feeling, and links to others; memories that matter get stronger with use, unused ones quietly fade (forgetting curve).
- Resonance-based recall: memories surface when your words resonate with them by meaning and significance, not keyword match — so only what resonates returns to context.
- Open source: pip install mnemos-continuity[mcp,embeddings]; the mnemos-mcp repo has 170 stars.
More from coding & agent
- OpenClaw ships 11.5/11.6: 9,954 PRs from 746 contributors focused on reliability — dean_rie · 2026-11-20
- Kaho AI learning tool update: notes that auto-draw diagrams, page-side Q&A — abhi9u · 2026-11-20
- When agent-built CLIs are too opaque, humans just let the agent run them — craigbalding · 2026-11-20
- Microsoft's free 18-lesson AI agents course hits 80.5k GitHub stars — victor_explore · 2026-11-20
- Self-hacking with abliterated models: biggest risks were infra side channels, not app code — cyb3rops · 2026-11-20
- Pocket City: a pixel-art Amsterdam canal simulation built entirely on real geodata — anselm · 2026-11-20