Identify and document LLM prompt-injection risks and tool-call abuse in AI systems.
npx skills add https://github.com/Xerrion/opencode --skill pentest-ai-redteam
Please help me install this Agent Skill. Skill: pentest-ai-redteam Source: https://github.com/Xerrion/opencode/tree/main/skills/pentest-ai-redteam Command: npx skills add https://github.com/Xerrion/opencode --skill pentest-ai-redteam
AI/LLM red-team testing frames and standardizes evaluation of prompt-injection vulnerabilities, tool-call abuse, and resilience against adversarial techniques.
Kick off a red-team engagement by selecting an AI system and following the playbook to enumerate injection vectors and produce reproducible evidence.
High-intent search queries and answers about installing and using this skill.
Identify LLM prompt-injection risks and tool-call abuse by applying adversarial probes and mapping test results to the OWASP LLM Top 10 and MITRE ATLAS frameworks for structured security assessments.
Evaluate RAG system resilience against indirect prompt injection by running standardized attack probes and recording deterministic transcripts with canaries to document instruction override and role confusion vulnerabilities.
Yes, you can assess ML classifiers and sandboxed environments by applying adversarial probe patterns and tool-call abuse scenarios to evaluate system resilience and map findings to MITRE ATLAS.
Document AI red-team evidence by imposing deterministic transcripts, canaries, and secure logging to the .pentest/evidence directory to ensure reproducible test scenarios and standardized vulnerability tracking.
Yes, AI red-team testing covers jailbreaking and system-prompt extraction by utilizing a comprehensive attack library that targets direct and indirect prompt injections, instruction overrides, and role confusion.
jinyimeng01Test AI and LLM applications for adversarial security weaknesses.
SnailSploitAssess AI and LLM systems for prompt injection and adversarial vulnerabilities.
pluginagentmarketplaceRun Python-based prompt-injection and adversarial security tests in CI/CD pipelines.
yogi100xIdentify AI system vulnerabilities through systematic adversarial testing.
