tickernelz/opencode-mem — explained in plain English
Analysis updated 2026-05-18
Let your AI coding assistant remember project architecture and past decisions between sessions.
Build an automatic profile of your coding preferences that the assistant reuses over time.
Browse and search everything the assistant has remembered through a local web dashboard.
| tickernelz/opencode-mem | affaan-m/agentshield | indalok/rzweb | |
|---|---|---|---|
| Stars | 711 | 719 | 723 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | easy | easy | easy |
| Complexity | 2/5 | 3/5 | 3/5 |
| Audience | vibe coder | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Installs with a single config line and downloads automatically on next OpenCode startup.
OpenCode Memory is a plugin for OpenCode, a coding agent tool, that gives AI coding assistants a persistent memory across sessions. Normally an AI assistant forgets everything once a conversation ends, but this plugin stores facts, decisions, and context in a local vector database on your own machine, so the assistant can recall past project details and your personal preferences the next time you work together. The plugin uses SQLite as its underlying data store, paired with a vector search library called USearch for fast lookups, and falls back to a simpler exact search method if USearch is not available. It automatically learns a profile of the user over time, keeps a combined timeline of memories and prompts, and includes a full web interface for browsing and managing what has been remembered. It also tries to avoid storing duplicate information and includes some built in privacy protections. To install it, you add opencode-mem to your OpenCode configuration file, and it downloads automatically the next time OpenCode starts. Once running, you can add memories, search past memories by keyword, view your learned profile, or list recent entries, either through simple commands or through the local web dashboard, which runs on port 4747 by default. For automatic memory capture, the plugin can use whatever AI provider you already have set up inside OpenCode, including Claude Pro or Max through OAuth, GitHub Copilot, or direct API keys, so you generally do not need to configure a separate key just for this plugin. There is also a manual fallback option if you prefer to point it at a specific provider like OpenAI directly. Settings such as storage location, memory limits, and how many past memories get injected into a new chat are all configurable in a separate settings file. The project is written in TypeScript, is open to contributions, and is released under the MIT license. It says it was inspired by an earlier project called opencode-supermemory.
A plugin that gives AI coding assistants persistent memory of your projects and preferences using a local database on your machine.
Mainly TypeScript. The stack also includes TypeScript, SQLite, USearch.
MIT license, so you can use, modify, and redistribute it freely, including commercially, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.