Analyze AI agents for tool interaction and dynamic attack path vulnerabilities.
npx skills add https://github.com/taoq-ai/ziran --skill ziran
Please help me install this Agent Skill. Skill: ziran Source: https://github.com/taoq-ai/ziran/tree/main Command: npx skills add https://github.com/taoq-ai/ziran --skill ziran
This Skill enables automated identification of security risks within AI agents by analyzing their capabilities, tools, and interactions.
Run the ZIRAN security analysis against your AI agent by providing its source code or API endpoint to identify potential security issues.
High-intent search queries and answers about installing and using this skill.
Detect vulnerabilities in AI agent tool chains by analyzing agent capabilities and tool interactions to identify dangerous sequences. The Skill constructs a capability graph to map tools and discover dynamic attack paths within multi-agent systems.
Yes, you can simulate attack campaigns against remote API AI agents by running multi-phase exploit scenarios. The Skill orchestrates multi-step attacks to evaluate agent safety and identify side-effects under varied conditions before deployment.
The best way to test for prompt injection in multi-agent systems is through automated security event detection. The Skill identifies unsafe behavior and prompt injections at both the text and execution levels during campaign simulation.
Attack simulation evaluates AI agent safety by running multi-step exploits against the target agent's capabilities. It requires graph analysis and detection modules to assess vulnerabilities arising from tool interactions comprehensively.
Yes, this vulnerability assessment works with proprietary AI agent source code. You provide the source code or API endpoint to run the security analysis and identify potential issues within your deployment pipeline.
You need tool chain mapping for AI security testing when you must automatically detect dangerous tool sequences. It constructs a capability graph to visualize and evaluate how interacting tools create dynamic attack paths.
Tai-ch0802Analyze AI agent skills for malicious code and prompt injection risks.
EojEdredScan code and configurations for security vulnerabilities and replay risks.
jassicsAutomate MASTG-based mobile penetration tests on Android and iOS apps.
AnsteorraPerform static and dynamic security audits on the KMP application.
