About Prompt Injection Scanner MCP - LLM Security Testing
Prompt Injection Scanner MCP - LLM Security Testing is an MCP server in the Scraping category: promptGuard Scan MCP helps SaaS teams test LLM applications for prompt injection, jailbreaks, indirect prompt injection, hidden instruction leakage, unsafe tool use, prompt security risks, and AI release-gate failures. Use it for prompt injection scanner, LLM security testing, AI red teaming, jailbreak detection, prompt security audit, LLM vulnerability scanner, ChatGPT security testing, and CI/CD safety checks for AI apps. The server provides read-only MCP tools for site profile, health status, pricing, analytics summary, conversion funnel, checkout status, search status, sitemap, robots, page usability, patrol reports, open issues, and recent issue updates. Customers purchase access on PromptGuard Scan and connect with a Bearer MCP token.
Install
git clone https://github.com/clauxel/prompt-injection-scanner-mcpThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Ask AI
Ask AI about Prompt Injection Scanner MCP - LLM Security Testing
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.