core
Verified Safeby opensumi
Overview
An extensible IDE framework for building web-based, desktop (Electron), and AI-native development environments, providing core UI components like message notifications and modal dialogues.
Installation
node packages/startup/dist-node/server/server.jsEnvironment Variables
- NODE_ENV
- PORT
- MY_WORKSPACE
- EXTENSION_HOST_ENTRY
- WATCHER_HOST_ENTRY
- WS_PATH
- WEBVIEW_HOST
- STATIC_SERVER_PATH
- KTLOG_SHOW_DEBUG
- PTY_PROXY_SOCK
Security Notes
The framework extensively uses `new Function()` for loading and activating extensions (e.g., in `ext.host.ts`, `worker.host.ts`, `webview-manager.ts`). While this is a common pattern for extensible IDEs like VSCode, it means the runtime executes code from extensions, requiring a high degree of trust in installed extensions. Webviews are sandboxed, but the `allow-same-origin` permission could be a concern if coupled with other vulnerabilities. Development/CLI server configurations (e.g., `tools/dev-tool/src/server.ts`, `tools/cli-engine/src/node/server.ts`) exhibit broad network access policies (`staticAllowOrigin: '*'`, open CORS), which are acceptable for local development but would be risky in a broadly exposed production environment. Sensitive `GITHUB_TOKEN` is used for changelog generation (development task) but is handled via environment variables. `keytar` is used for secure credential storage (good practice).
Similar Servers
mcp-use
A comprehensive framework for building full-stack Model Context Protocol (MCP) applications, including AI agents, MCP servers with UI widgets, and integrated debugging tools in both Python and TypeScript.
agents
This MCP (Model Context Protocol) Server acts as a proxy for the Inkeep Agent Framework's Management API. It exposes administrative functionalities (e.g., CRUD operations for agents, projects, tools, and credentials) via the MCP protocol, allowing other clients or agents to interact with the Inkeep Management API through a standardized interface.
arcade-mcp
Provides a framework and pre-built toolkits for integrating Large Language Models (LLMs) with various external services and databases, enabling AI agents to interact with the real world.
claude-prompts
This server provides a hot-reloadable prompt engine with chains, quality gates, and structured reasoning for AI assistants, enhancing control over Claude's behavior in prompt workflows.