"Firewalla" matching MCP servers:

Enables autonomous agents and MCP-compatible clients to route inputs through a deterministic firewall that flags and sanitizes prompt injection, jailbreaks, system prompt exfiltration, and tool hijacking. It supports JSON-RPC stdio integration for tools like Claude Desktop, Cursor, and multi-agent frameworks.

A security filter that blocks dangerous code patterns by comparing normalized structural syntax trees against a blacklist of known threats using vector embeddings. It acts as a gatekeeper to prevent malicious code execution by identifying dangerous structures regardless of specific identifiers or literals.

Enables AI agent tool calls to be intercepted and judged before execution, allowing, asking a human, or blocking based on calibrated risk and policy.

A transparent proxy that intercepts high-risk tool calls and requires OTP-based human approval before they can be executed. It acts as a configurable circuit breaker between AI agents and target MCP servers to prevent unauthorized or dangerous actions.

A policy-based security layer for AI agents and MCP tools that evaluates tool requests against security policies before allowing, denying, or requiring human approval for execution.