F
licenseNot graded
qualityNot graded
maintenanceEnables MCP-compatible clients to deterministically neutralize adversarial prompt injections and jailbreak overrides with zero-dependency, sub-millisecond heuristic analysis.
1
-
genpark-agent-prompt-injection-sanitizer-firewall-skill
Officialby Alpha-Park
Related Servers
Alternatives to genpark-agent-prompt-injection-sanitizer-firewall-skill
No user-submitted related servers found.
Related Servers
- FlicenseNot gradedqualityBmaintenanceEnables deterministic detection and neutralization of adversarial prompt injections and override attempts in AI agent workflows via a zero-dependency MCP server, providing structured telemetry and low-latency validation.8-
- FlicenseNot gradedqualityBmaintenanceEnables 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.7-
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients and autonomous agents to screen incoming prompts for system overrides, delimiter breakouts, and encoded payload evasion, helping block prompt injection and jailbreak attempts before execution.8MIT
- AlicenseAqualityAmaintenanceEnables sub-millisecond pre-execution safety filtering for AI agent tool calls, blocking destructive shell commands, dangerous SQL mutations, credential access, SSRF, and scope-creep or prompt-injection patterns before they execute, with optional fail-closed transparent proxy wrapping for any MCP server.669 npm12Apache 2.0
- AlicenseNot gradedqualityNot gradedmaintenanceEnables real-time injection fuzzing, exfiltration canary detection, least-privilege tool sandboxing, and cryptographic audit proofs for untrusted prompts, RAG context, and agent actions over MCP.7MIT
