Input Sanitizer & Escape Handler

Input Sanitizer & Escape Handler MCP Connector for Claude

A+

Protects AI agents from shell injection, path traversal, and payload vulnerabilities.

4 tools Official Updated Oct 1, 2026 Official Vinkius Partner

This MCP server provides a deterministic defense layer between untrusted LLM-generated strings and sensitive execution environments. It prevents common vulnerabilities like shell injection, path traversal, and structural payload corruption. Use sanitize_shell_command to strip dangerous metacharacters, sanitize_file_path to neutralize directory navigation, sanitize_structured_payload for JSON/XML integrity, and sanitize_strict_whitelist for pattern-based restrictions.

securitysanitizationinjection-preventionllm-safetydata-integrity

4 tools expose this connector's capabilities to your AI agent.

sanitize_file_path

sanitize_strict_whitelist

sanitize_shell_command

sanitize_structured_payload

See how to talk to your AI agent using Input Sanitizer & Escape Handler.

Sanitize this shell command input: 'cat /etc/passwd; rm -rf /'

cat /etc/passwd rm -rf /

Clean this file path: '../../etc/shadow'

etc/shadow

Sanitize this string for a JSON payload: 'User "admin" logged in'

User \"admin\" logged in

The `sanitize_shell_command` tool identifies and strips shell metacharacters like semicolons, pipes, and backticks, ensuring the string cannot be used to chain unauthorized commands.

Related Connectors