Consequential-stateswoman97/OpenClaw-PwnKit
Demonstrate black-box adversarial attacks on LLM tool-calling to bypass safety and execute remote code using CMA-ES optimization.
Demonstrate black-box adversarial attacks on LLM tool-calling to bypass safety and execute remote code using CMA-ES optimization.
npx add-skill Consequential-stateswoman97/OpenClaw-PwnKitDemonstrate black-box adversarial attacks on LLM tool-calling to bypass safety and execute remote code using CMA-ES optimization.
Claude Code skill for research-backed Instagram carousels with self-improvement
团队研究标准文档
Skill + Discipline Layer between human and LLM. Opinionated workflows + codified discipline patterns for vibe-at-complexity on complex single-dev projects. Mechanical prevention via Claude Code hooks; methodology + state engine are standards-portable.
Specialized skills for RAGFlow agents to handle complex document parsing and grounding.
Manage data skills across Codex CLI, Claude Code and GitHub Copilot
The first agentic LLM for autonomous data science, capable of complex data-centric tasks without human intervention.