| FMRS | 50 / 100 · D | 57 / 100 · C | 55 / 100 · C |
|---|
| Reliability | 5 / 20 | 8 / 20 | 9 / 20 |
|---|
| Security and permissions | 10 / 20 | 10 / 20 | 8 / 20 |
|---|
| Maintenance | 13 / 20 | 16 / 20 | 13 / 20 |
|---|
| Documentation | 14 / 20 | 11 / 20 | 12 / 20 |
|---|
| Setup experience | 8 / 20 | 12 / 20 | 13 / 20 |
|---|
| Best for | - Users who want a desktop-grade AI agent that can directly operate local files, terminal, and browser workflows
- Users who want to remotely trigger desktop Agent tasks via WeChat, Feishu, DingTalk, Telegram, or similar IM channels
- Users needing office automation for documents, spreadsheets, slides, video, and web research
- Developers comfortable with Electron/React/TypeScript who want to build from source or use official installers
| - Existing Mindwtr users who want AI assistance with task capture and processing
- Developers or power users comfortable configuring a local stdio MCP server
- Self-hosters running a Mindwtr Cloud instance who need read-only querying
| - Teams needing a unified SQL interface across cloud and SaaS providers
- Users who want AI agents to access cloud resources in a structured way
- Cloud operations, security, FinOps, and infrastructure-as-code workflows
|
|---|
| Not for | - Users who need an MCP server that a standard MCP client (e.g. Claude Desktop) can connect to directly — no such server manifest or tool definitions are provided in this repository
- Users needing a headless or server-side deployment — this project only ships as a macOS/Windows desktop application
- Users who want a narrow, single-purpose tool integration without granting broad file, terminal, and network permissions
| - Users without Mindwtr installed or without a local SQLite database
- Teams needing multi-user collaborative writes via Cloud (Cloud mode is read-only)
- Non-technical users unfamiliar with npm package installs and environment variables
| - Users who only need local file or browser automation
- Environments that cannot grant cloud or SaaS API access
- Users expecting a tool inventory with individually named MCP tools, which the source does not provide
|
|---|
| Required permissions | - Local file system read/write
- Terminal command execution
- Browser automation
- Network access
- IM channel binding and messaging (WeChat, WeCom, DingTalk, Feishu/Lark, QQ, Telegram, Discord, NetEase IM, NetEase Bee, POPO, email)
- Local SQLite storage (lobsterai.sqlite under Electron userData)
| - Read/write access to the local Mindwtr SQLite database file (path set via MINDWTR_DB_PATH)
- Network access to a self-hosted Mindwtr Cloud URL when Cloud mode is configured (read-only)
- A bearer token credential for Cloud API access
| - Provider authentication must be supplied through --auth or the client configuration
- Effective permissions depend on the provider credentials and requested operations; the server supports querying and provisioning
- GitHub Actions defaults to read_only mode, which is intended as the safe CI default
|
|---|
| Risks and side effects | - The app requires broad, sensitive permissions (file operations, terminal, network); the README notes sensitive actions are permission-gated and logged, but users must still review approval prompts carefully to avoid data loss or exposure
- Remote control via IM channels means commands can trigger desktop actions from linked messaging accounts — compromise of a bound account/channel could lead to unauthorized local execution
- Packaging bundles the OpenClaw runtime and, on Windows, a portable Python runtime, expanding the local attack surface
- This project is not maintained by the MCP protocol owner and is not a conventional MCP server; anyone wanting standard MCP-protocol integration should verify exposed behavior directly, as the source material gives no such details
| - The server has write access to the local database, so erroneous AI-issued changes could corrupt or pollute local task data
- Storing the Cloud bearer token insecurely in environment variables risks credential exposure
- Task and project data is exposed to whichever AI model is connected, with no additional sandboxing on scope
| - Provisioning or modifying resources can cause real infrastructure changes
- Overprivileged credentials can broaden the impact across cloud or SaaS resources
- AI-generated SQL or API operations should be reviewed before execution
- Cross-cloud operations may create cost, compliance, and security consequences
|
|---|
| Supported clients | | Claude Desktop | Claude, VS Code, Cursor |
|---|
| Tools | 0 | 0 | 0 |
|---|