io.github.CSOAI-ORG/agent-prompt-injection-firewall-mcp

Official

by CSOAI-ORG · Python

The WAF for agents. Pattern-based + heuristic firewall scans prompts, RAG documents, tool argume...

Blocks prompt injection attacks on prompts, RAG documents, and tool arguments.

Topics

mcpmodel-context-protocolagentai-agentsai-complianceai-governanceai-securityclaudecompliancemcp-servermeokmeok-ai-labsmit-licenseopen-sourceowaspowasp-llmprompt-injectionsecuritycsoai

Install

claude_desktop_config.json

{
  "mcpServers": {
    "agent-prompt-injection-firewall-mcp": {
      "command": "uvx",
      "args": [
        "agent-prompt-injection-firewall-mcp"
      ]
    }
  }
}