Launching today

Humanbound

Launching today

Open-source security testing for AI agents and chatbots

0 followers

Open-source security testing for AI agents and chatbots

0 followers

Visit website

Humanbound continuously tests AI agents and chatbots with adversarial attacks, tags findings to OWASP LLM and Agentic Top 10, and scores security posture.

Free Options

Launch tags:Productivity

Launch Team

Hunter

📌

Humanbound is an open-source security testing platform that continuously tests AI agents and chatbots against adversarial attacks. Install via pip, point it at your agent's endpoint, and get an A-F security posture score. Features include multi-turn adaptive testing for prompt injection, jailbreaks, goal hijacking, tool misuse, and data exfiltration. Findings map to OWASP LLM and Agentic Top 10, MITRE ATLAS, NIST AI RMF, and regulatory frameworks. Includes pytest CI/CD plugin, custom firewall to block discovered attacks, and free Community dashboard. Bring your own LLM model. Apache 2.0 licensed, free for commercial use.

Report

19d ago