MemOS
MemTensor/MemOS
A memory operating system for LLMs and agents.
MemOS treats memory as an OS-level resource — scheduling, storing and evolving different memory types (parametric, activation, plaintext) with hybrid retrieval. A research-born take on industrializing agent memory.
In the news
First 'Harness Engineering' survey lands on arXiv
'Harness Engineering: Anatomy, Architecture, and Evolution of Coding Agents' surveys the field — loops, tool interfaces, context management, memory and recovery — giving the emerging discipline its first comprehensive academic map.
HarnessTax: harness choice swings agent cost more than model choice
Arena.ai's HarnessTax study tests 21 model–harness combinations across Claude Code, Codex CLI and Pi, finding harness choice can significantly change cost even at similar task success rates. Weeks earlier, SWE-bench Pro analysis showed swapping harnesses moved pass@1 more than model upgrades (23%→52% on GLM-5.2) — and that harness rankings barely transfer across models.
Related projects
Hermes Agent
NousResearch/hermes-agent
The agent that grows with you.
Nous Research's self-improving personal agent: a learning loop turns experience into reusable skills, builds a persistent user model across sessions, and checkpoints state to disk with rollback — lean enough for a $5 VPS and driven from the chat apps you already use.
claude-mem
thedotmack/claude-mem
Persistent context across sessions for every agent.
claude-mem captures everything an agent does during a session, AI-compresses it, and injects the relevant context into future sessions — session-to-session memory as a plugin for Claude Code, Codex, OpenClaw, Gemini, Copilot and more.
Headroom
headroomlabs-ai/headroom
Context compression for coding agents.
Headroom compresses tool outputs, logs, files and RAG chunks before they reach the LLM — 20% fewer tokens for coding agents and 60–95% for other workloads. Context engineering as infrastructure: keeping the window usable as harnesses grow.