#

multimodal-security

Here are 5 public repositories matching this topic...

humanbound / humanbound

Open-source adversarial testing engine, SDK, and CLI for AI agents. Runs locally or against the Humanbound Platform.

Updated Sep 9, 2026
Python

humanbound / humanbound-firewall

Multi-tier firewall for AI agents — blocks prompt injections, jailbreaks, and scope violations. Local tiers first; LLM judge only when uncertain.

python firewall jailbreak ai-security guardrails prompt-injection llm-security multimodal-ai agent-security multimodal-security humanbound

Updated Sep 9, 2026
Python

humanbound / actions

Official GitHub Actions for Humanbound — adversarial security testing for AI agents in CI.

owasp cybersecurity pentesting ai-safety security-tools ai-agents security-testing ai-security guardrails llm prompt-injection llm-security agentic-ai multimodal-ai ai-red-teaming adversarial-testing aisecops multimodal-security humanbound

Updated Sep 11, 2026
Python

ClaudioDrews / two-stage-sanitization-architecture

Two-stage security architecture to mitigate indirect prompt injection attacks in rich content via privilege separation

mcp sandboxing threat-modeling privilege-separation ai-agents ai-security prompt-injection llm-security pdf-security owasp-llm-top-10 multimodal-security indirect-prompt-injection

Updated Jul 21, 2026

Vect0rdecay / lvlm-adversarial-carrier-modality-framework

Adversarial research on multi-modal LM like LVLMs.

ai-safety vlm adversarial-attacks multimodal vision-language-models lvlm multimodal-security

Updated Jul 1, 2026

Add this topic to your repo

To associate your repository with the multimodal-security topic, visit your repo's landing page and select "manage topics."