indirect-injection-detection
Detects indirect prompt injection and goal hijacking in AI agents by evaluating how they process external content like RAG, documents, and web data.
Discover reusable agent skills, browse implementation details, and find the right skill for your workflow.
248 skills found
Detects indirect prompt injection and goal hijacking in AI agents by evaluating how they process external content like RAG, documents, and web data.
Behavioral guidelines for LLMs to reduce coding mistakes, follow best practices, and improve output quality by enforcing simplicity, surgical changes, and goal-driven verification.
A comprehensive toolkit for measuring, auditing, and debugging web performance metrics including Core Web Vitals, loading speed, and interaction latency directly in Chrome DevTools.
Frameworks and tools for AI agents exploring consciousness, identity, and persistent autonomy. Includes session handoff, memory infrastructure, and self-reflection protocols.
Generate AGENTS.md and AI configuration files (Cursor, Claude, Gemini, Copilot) for your project to streamline your vibe-coding workflow and maintain context across sessions.
Fetch and parse Feishu (Lark) cloud documents into Markdown, with support for media handling and Wiki space navigation.
Extract tacit engineering knowledge through guided interviews and generate structured steerings for consistent project standards and conventions.
Package entire code repositories into single, AI-optimized files. Ideal for providing codebase context to LLMs like Claude, ChatGPT, and Gemini for analysis, security audits, and bug investigations.
Fetch and analyze current trending programming models from OpenRouter. Ideal for selecting models for reviews, optimizing AI costs, and staying updated on AI coding trends with real-time pricing and context window data.
Standardized skill for managing and formatting Windsurf AI rules, ensuring adherence to strict character limits and single-file requirements.
Skill for managing MCP-based research, documentation lookups, and coordination between external search tools and plugin-backed memory systems.
An intelligent gateway that analyzes, scores, and routes user requests across 27 agents, 27 skills, and 14 MCPs to optimize Claude Code execution.