OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Always Accompany

Always Accompany is an open-source JavaScript AI agent framework that combines companionship, chat, coding and work modes through shared memory, configurable context, specialized tools and project isolation.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Always Accompany — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when conversation, coding and work need to share an agent context, or when you want to experiment with configurable memory, attention, project isolation and specialized modes in a local framework.

Who it is for

  • Developers exploring multi-purpose AI agents and context management
  • Users seeking one local interface for conversation, coding and work tasks
  • Experimenters working with plugins, MCP tooling, prompt caching or specialized agent modes

Documented uses

  • Hold long-term chat and roleplay using imported character cards, presets and worldbooks
  • Read and modify project files and run commands as a local agent workbench
  • Retrieve relevant long-term material for current turns through layered memory and context handling
  • Separate working context between projects or user tasks
  • Extend the framework with JavaScript, Python or standalone-program plugins
Before you choose

Requirements & limits

The repository is public, accessible through GitHub and licensed under AGPL-3.0. The documented first-chat setup requires a working AI API, including its service URL, API key and model. The project provides launcher scripts for Windows and Linux/macOS; the launcher can download Deno when needed.

  • A working AI API is required for first chat use.
  • The connected model affects capability and cost.
  • Project isolation is explicitly described, but implementation details are not provided.
  • MCP is included in the toolchain, while a specific MCP server setup is not shown.
  • Optional retrieval, voice and desktop-companion features may require additional configuration or environment support.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Always Accompany for a first chat

Prepare the local interface, connect an AI service and begin a character-based chat.

Before you start

  • A working AI API
  • An API service URL, key and model
  • Git and the project repository
  • Ability to write simple prompts

Steps

  1. Clone the repository and enter its directory.

  2. Run run.bat on Windows, or make run.sh executable and run it on Linux or macOS.

  3. Wait for the runtime, dependencies and local data initialization to complete; open the local interface if the browser does not open automatically.

  4. Choose the interface language, enter the AI service URL, API key and model, then save the connection.

  5. Select or import a character card to start chatting.

Expected outcome

The local interface is configured with an AI service and a selected character for chat.

Guide limitations

  • A working AI API is required.
  • Model capability and cost depend on the connected service.
  • First launch can take longer while the runtime and dependencies are downloaded.
  • The guide documents initial chat setup, not a successful execution test.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 October 2026
Original website title
beilusaiying/always-accompany · Official homepage title · retained in its source language
How we review sourcesSuggest a correction