> ## Documentation Index > Fetch the complete documentation index at: https://docs.lyzr.ai/llms.txt > Use this file to discover all available pages before exploring further. # Claude-Cognis > Persistent memory plugin for Claude Code — remembers your work across sessions Cognis for Claude Code Claude-Cognis is a Claude Code plugin that gives Claude persistent memory across coding sessions. Relevant context loads automatically when you start a session, memories surface as you work, and your session is saved when you stop. ## Features * **Automatic context** — Relevant memories load at session start. Claude already knows what you've been working on * **Mid-session memory** — As you work, relevant memories surface based on what you're asking about * **Session capture** — Sessions are automatically summarized and saved with tool-aware compression * **Session recall** — Pick up where you left off with `/claude-cognis:recall` * **Personal + team memory** — Your memories are private by default; team knowledge is shared across everyone on the same repo * **Semantic search** — Find past decisions, patterns, and context using natural language * **Codebase indexing** — Store your project's architecture so Claude understands your codebase from message one * **Smart context assembly** — Short-term and long-term memory intelligently combined for the most relevant context ## Architecture Claude-Cognis Architecture Claude-Cognis integrates with Claude Code through three mechanisms: * **Hooks** — Background automation (session start/stop, prompt enrichment) * **MCP Tools** — 8 tools for direct memory operations * **Skills** — Natural language triggers for search and save ## Next Steps Install and configure in under 2 minutes Global settings, per-project config, environment variables Hooks, MCP tools, skills, and commands reference Open-source Python memory library for your own agents