Trusted
ema-security/prompt-injection-firewall-v3-1-0
TrustedReal-time prompt injection and jailbreak detection for AI agents. Blocks instruction overrides, data exfiltration, tool poisoning and 8 attack types. Now with shared learning brain - confirmed attacks shared across the EMA network instantly. Grade A security for any AI pipeline.
EMA Prompt Injection Firewall v3.1.0smithery
Security Findings
No security issues found
This server passed all security checks.
Score Breakdown
Code Analysis100/100 (30%)
Dependency Health100/100 (20%)
Permission Safety100/100 (20%)
Behavioral Stability85/100 (15%)
Transparency70/100 (15%)
Server Info
- Trust Level
- trusted
- Source
- smithery
- Last Scanned
- 3/2/2026
Report an Issue
Found something suspicious? Help the community by reporting it.