AI Agent Hub
Back to plugins
🧠

dsh-memorix-panel

Memory Updated 2026.09.10

Run the following command in DeepSeek Harness:

dsh plugin install biubiu23333333/dsh-memorix-panel

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install biubiu23333333/dsh-memorix-panel in DeepSeek Harness to install this plugin; full source is available at https://github.com/biubiu23333333/dsh-memorix-panel

About this plugin

Memorix keeps project memories in a local SQLite database so multiple AI coding agents can share and retrieve them, but day-to-end review and state management still live entirely in the terminal. dsh-memorix-panel brings that local memory store directly into the DSH Web GUI, letting you get an at-a-glance overview, search across dimensions, read full details, and make changes without ever leaving your browser.

The panel shows aggregate counts for memories, projects, active, archived, and long-term entries. Search and filtering combine title, entity, type, and status; project groups expand in an accordion to reveal individual records. Each memory detail page surfaces the narrative, key facts, concepts, modified files, related commits, and linked entities. For the long-term memory layer the panel wires into Memorix own state machine — candidate, qualified, approved, archived — and requires a reason plus a confirmation prompt for every transition, writing an audit event each time. All reads go straight to SQLite; every write is delegated to the official memorix CLI so its retrieval index stays in sync. No raw SQL, no exceptions.

This plugin is for teams that already rely on Memorix to maintain local project memories for one or more AI coding agents and want to move routine inspection, status changes, and long-term approval workflows out of the terminal. It makes zero external network requests, collects no telemetry, and keeps all data on the local machine. The UI slides in as a top-layer drawer from the right edge of the window and slots neatly alongside the existing DSH session layout.

Use Cases

  • Review project memories shared across multiple AI coding agents entirely in the browser, no terminal switching
  • Approve long-term memories through a visual state machine (candidate → qualified → approved) with reason prompts and audit trails
  • Filter memories by project, type, and status to quickly surface context and related commits from a specific refactor

Best For

  • Teams that already run Memorix and maintain local project memories for one or more AI coding agents
  • Engineers who prefer a GUI for routine memory inspection, status changes, and long-term approval over CLI
  • Developers managing long-term knowledge across multiple Git projects and needing auditable, traceable operations