A secure, production-ready MCP server for managing Virtual Private Servers via SSH
Server has 4 tools with complete input schemas and descriptions, but quality is inconsistent. Tool names follow verb_noun convention (exec_command, read_file, write_file, upload_file). Descriptions are present but generic (avg ~80 chars, below 194-char baseline). Parameters have types and descriptions, but lack constraints (enums, ranges, patterns). No output schemas documented. Error handling guidance absent. Security concerns: exec_command accepts arbitrary shell commands with minimal validation; no mention of permission gates or audit trails. File operations lack path traversal protection details.
Execute shell commands on VPS servers with security validation
Read file content from VPS servers via SFTP
Upload file from local system to VPS server
Write file content to VPS servers via SFTP
exec_command lacks output schema documentation and error recovery guidance. LLM cannot predict return structure or know how to handle command failures.
No input validation or constraint documentation for command parameter. Descriptions lack format/range guidance (e.g., max command length, forbidden patterns). LLMs may pass malicious payloads without detection.
File operation tools (read_file, write_file, upload_file) lack path traversal protection documentation. No mention of allowed directories, symlink handling, or validation rules in descriptions.
Tool descriptions are generic and under 100 chars (baseline 194). 'Execute shell commands on VPS servers with security validation' does not explain WHEN to use this vs alternatives, WHAT it returns, or prerequisites.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | <=2025-11-25 | v2 |
No permission gates or audit trail documentation. Tools modify/access remote systems but lack descriptions of permission checks, logging, or compliance controls.