M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
A

AI as Workspace

Community
Not a standard server
Cross-platform AI chat client with MCP support and local-first storage.

This is not a typical connect-and-use MCP server; read the source repository before using it.

How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #290 of 438
Stars
★ 1.8k Very popular
Transport
stdio (local process) · SSE · Streamable HTTP
Runtime
Prebuilt binary · Node.js
Credentials
No credential needed
License
BSD-3-Clause
Last commit
47FMRS · D

AI as Workspace is a cross-platform, local-first AI chat client that acts as an MCP client supporting Tools, Prompts and Resources over STDIO and HTTP, with features such as multiple workspaces, Artifacts, a plugin system and dynamic prompts. Note that the repository is marked deprecated and succeeded by Nya AI, so new users should consider the successor project first.

Strongest · Documentation 12/20 Weakest · Maintenance 6/20

Reliability
8/20
Security and permissions
11/20
Maintenance
6/20
Documentation
12/20
Setup experience
10/20
Why each score
Reliability 8/20
This repository is an MCP client (AI as Workspace), not an MCP server; no server manifest is provided at the repository root, so a static review cannot verify MCP initialization, the tools/list handshake, or whether declared tools match real behavior. The README only states support for Tools/Prompts/Resources over STDIO and HTTP, with no committed tests covering those key paths and no visible CI workflow evidence, which caps reliability at 12 under static calibration. The README explicitly marks the project deprecated in favor of Nya AI, further weakening confidence in clean startup. The feature set and stack (Quasar/Vue, local-first) are plausible and the happy path seems workable, but absent reproducible execution evidence and edge/error-handling detail, it scores 8.
Security and permissions 11/20
No signs of malware, credential theft, or covert exfiltration were found, and no real tokens or secrets appear in examples, so no red line is triggered. However, static evidence shows multiple external-network and side-effect paths: STDIO and HTTP remote MCP servers, marketplace and Gradio plugin installation, SearXNG web search and URL crawling, and cloud sync. The README and docs do not disclose a least-privilege model, confirmation flows for external MCP calls, plugin sandbox boundaries, or data-flow details (encryption and location of synced local-first data). Artifact read/write permission control is mentioned but lacks concrete scope and confirmation detail. Main risks are visible while permissions, confirmation, and scoping are incomplete, so slightly above the 10 anchor, scoring 11.
Maintenance 6/20
A clear open-source license (BSD-3-Clause) is present, and stars and issue counts indicate prior community use, but the README's first line declares the project deprecated and succeeded by Nya AI with a migration guide, indicating maintenance has ended, with no ongoing issue fixes or dependency updates. No release cadence, security-response channel, or governance structure is evidenced. Against anchors of 0 for archived/abandoned and 10 for active with governance gaps, this sits on the abandoned side but retains a license and a history of activity, scoring 6.
Documentation 12/20
The README plus docs-site links (usage, self-hosting, MCP, Artifacts, plugins, prompt variables, migration) form a reasonably layered set covering feature overview, dependency install (pnpm i, quasar dev/build), platform support, and the plugin ecosystem, making it usable documentation. It lacks MCP client connection config examples, external-server auth methods, tool parameter and limit descriptions, cost and rate-limit notes, and a troubleshooting section. No contradictions between the README and claimed capabilities, matching the 10 anchor. Detail is adequate but hidden assumptions and troubleshooting gaps remain, scoring 12.
Setup experience 10/20
The install path is given via dependency install and build commands (pnpm i, quasar dev, quasar build, build -m pwa), with a web version, download page, Microsoft Store listing, and self-hosting guide. Client configuration examples for mainstream clients are absent (as a client rather than a server, there is no example of wiring it into Claude/Cursor). Steps depend on a Node/pnpm runtime and the Quasar toolchain, a workable but manual flow requiring self-managed build and deployment. Under static calibration with no execution evidence, setup is capped at 15; given clear multi-platform distribution channels but missing connection config examples, it scores 10.

Static review · not runListed 2026-09-10

Read the FMRS scoring method →

Fit and risk

Best for

  • Users who want a consistent AI chat experience across multiple operating systems
  • Users who prefer local-first conversation storage with optional cloud sync
  • Users who want to consume tools, prompts and resources through the MCP protocol
  • Advanced users who value plugins and dynamic prompt variables for customization

Not for

  • Users looking for an actively maintained project (this one is deprecated; migration to Nya AI is recommended)
  • Users who only need a simple command-line tool without a graphical interface
  • Enterprise-grade centralized control or strict compliance auditing scenarios

Required permissions

  • Network access: connect to configured AI providers (OpenAI, Anthropic, Google, DeepSeek, xAI, Azure, etc.)
  • Local storage/file system: store data and file attachments locally first
  • Process execution: launch local MCP server processes over STDIO
  • Cloud account: used for cross-device data synchronization after login

Risks and side effects

  • The project is deprecated and may no longer receive security updates or feature fixes
  • Local MCP servers launched over STDIO inherit client process permissions; only add servers you trust
  • Cloud sync uploads data to a server, creating privacy and data leakage risks
  • Plugins and third-party services may access conversation content, risking data exposure when untrusted
  • Connecting to third-party AI providers sends conversation content to those providers

Setup

Before you start

Runtime：Prebuilt binary · Node.js

Source install steps:

  1. Run pnpm i to install dependencies.
  2. Run quasar dev to start the app in development mode (hot reload, error reporting).
  3. Run quasar build to build the production SPA, or quasar build -m pwa for PWA.
  4. For self-hosting, refer to the official self-hosting guide.

MCP configuration: install MCP-type plugins from the plugin marketplace, or manually add MCP servers using STDIO or HTTP connection methods.

Check that it works

This project is deprecated (succeeded by Nya AI) and is an MCP client rather than an MCP server. Verify by installing an MCP plugin from its plugin marketplace and checking that the server appears in the client's MCP server list.

Troubleshooting

  1. If an MCP server fails to connect, verify the STDIO command path or the HTTP address is correctly configured
  2. If local data is not syncing, confirm you are logged in and the sync service is reachable
  3. If the build fails, confirm pnpm dependencies are installed and a compatible Node.js version is used
  4. If behavior is abnormal on a platform, try updating to the latest release or using the Web/PWA version
  5. If a plugin does not work, check it is correctly installed and enabled from the plugin marketplace

Use cases

Everyday cross-platform AI conversations with multiple workspaces
Using it as an MCP client to connect local or remote MCP servers with Tools, Prompts and Resources
Converting assistant responses into editable, version-controlled Artifacts
Extending capabilities via plugins such as calculator, document parsing, video parsing and image generation
Working offline-first and syncing data across devices after login

Overview

AI as Workspace is a full-featured yet lightweight AI chat client available across Web, PWA, Windows, Linux, macOS and Android. It offers multiple workspaces, a plugin system, Artifacts and dynamic prompts, and acts as an MCP client supporting MCP Tools, Prompts and Resources over STDIO and HTTP connections. Data is stored locally first, with optional cloud synchronization after login for cross-device use. The project is marked deprecated and succeeded by Nya AI.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 63.3k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision e9e1c55d0b8e Data synced 2026-10-19 Read the FMRS scoring method