tickernelz/opencode-mem

OpenCode plugin that gives coding agents persistent memory using local vector database

What it solves

OpenCode Memory provides a persistent, long-term memory system for AI coding agents. It prevents the loss of technical context, project-specific decisions, and user preferences across different chat sessions, ensuring that AI agents don't forget important details as a project evolves.

How it works

The system uses a local vector database (Turso/libSQL) to store and retrieve memories. It employs embedding models (via Hugging Face transformers or OpenAI-compatible APIs) to convert text into vectors for similarity search.

Key mechanisms include:

  • Auto-capture: A background AI process that automatically summarizes and saves technical context when a session goes idle.
  • User Profile: A cross-project summary that learns the user's habits and preferences over time.
  • Memory Injection: Relevant memories are automatically injected into the AI's context during conversations.
  • Manual Control: A memory tool allowing users to explicitly add, search, or migrate memories.

Who it’s for

Developers using OpenCode AI agents who want their AI to retain project-specific knowledge and personal coding preferences across multiple sessions and projects.

Highlights

  • Local-first storage: Uses embedded Turso/libSQL with native vector search for privacy and speed.
  • Automatic learning: Extracts memories and auto-updates a user profile without requiring manual input.
  • Web UI: A dedicated interface at http://127.0.0.1:4747 to browse, inspect, and manage memories.
  • Flexible Embeddings: Supports 12+ local embedding models or remote OpenAI-compatible endpoints.
  • Workspace Support: Handles multi-repo workspaces using a .opencode-mem-project marker file to share memory across nested repositories.

Related

  • Project
    supermemoryai/opencode-supermemory

    A plugin for OpenCode that adds persistent, long-term memory to AI agents, allowing them to remember project details and user preferences across sessions.

  • Project
    joshuadavidthomas/opencode-agent-memory

    A plugin for OpenCode that gives AI agents persistent, self-editable memory blocks and a semantic journal to maintain context across sessions.

  • Project
    Lyellr88/marm-memory

    A persistent local memory server for AI agents using the Model Context Protocol (MCP), providing shared long-term memory, code graphs, and concept graphs across sessions.

  • Project
    rohitg00/agentmemory

    A persistent memory server for AI coding agents that automatically captures and retrieves project context across sessions using hybrid search.

  • Project
    CodeAbra/iai-personal-memory-engine

    A local, MCP-compatible personal memory engine for AI coding assistants that automatically captures and recalls session history to provide long-term context and reduce token usage.