$ loading_
Local-first memory for Claude Code and Codex with capture, distillation, and recall.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "io.github.majiayu000/remem" MCP server from askskill: Run: claude mcp add 'io-github-majiayu000-remem' -- npx -y @remem-ai/remem
Developers can use it across multiple Claude Code or Codex sessions to save and recall project context, reducing repeated explanations. It fits long-running tasks involving decisions, constraints, and todos.
When users collaborate continuously with coding assistants, this tool can capture and distill key information from the conversation. That makes it easier to restore previous working context in later sessions.
Users who prefer a local-first workflow can use it to manage and recall memory stored in SQLite. It suits people who value local data retention and efficient context recovery.
It provides local-first automatic memory for Claude Code and Codex, including capture, distillation, and recall. The description says its memory is stored in SQLite.
Based on the provided description, memory is stored in SQLite. For the exact schema and storage location, see the source repository.
From the description, it emphasizes automatic capture, distillation, and recall rather than only keeping raw chat logs. For any additional organization features, see the source repository.
Add persistent database-backed memory to Claude Code.
Give Claude Code cross-session long-term memory with SQLite full-text search.
Give Claude Code cross-machine, file-first memory with synced full-text search.
Search local AI coding chat history and retrieve compact evidence fast.
Give Claude Code persistent memory, checkpoints, and project sync across sessions.
Provide local-first multilingual memory for Codex, Claude Code, and MCP clients.