The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions found attempting to override agent behavior or bypass safety filters. The content is strictly instructional documentation for web development.- [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials or unauthorized network operations were found. API interaction examples use generic placeholders (e.g., '/api/data') consistent with standard web development practices.- [Obfuscation] (SAFE): No Base64 encoding, zero-width characters, or homoglyphs were detected. The markdown and code blocks are clear and legible.- [Unverifiable Dependencies] (LOW): The files reference standard, industry-recognized libraries such as @mui/material, @tanstack/react-query, and zod. These are used as examples for development standards rather than being executed by the skill itself.- [Privilege Escalation] (SAFE): No commands for administrative access or permission modification were found.- [Persistence Mechanisms] (SAFE): No attempts to modify shell profiles, scheduled tasks, or startup scripts.- [Indirect Prompt Injection] (LOW): As documentation for an AI agent, the skill provides patterns for handling external data. It actively encourages security best practices by recommending Zod for schema validation and strict TypeScript typing for data ingestion points.- [Dynamic Execution] (SAFE): No use of eval(), unsafe deserialization (pickle/yaml), or runtime code generation.

frontend-dev-guidelines