The Agent Skills Directory

[SAFE]: The skill serves as a static knowledge base and architectural guide for designing AI agents. It does not contain any executable code or instructions that perform network operations, file system modifications, or credential access.\n- [SAFE]: Security best practices are explicitly documented within the reference files. For instance, references/shared-workspace-architecture.md provides guidance on scoping workspaces to prevent path traversal and protecting sensitive files like .env or credentials.json.\n- [SAFE]: The reference files contain various code examples for languages like TypeScript and Swift. These are presented within markdown code blocks for educational purposes and are not used as payloads or hidden commands.\n- [SAFE]: No malicious patterns such as prompt injection, obfuscation, or data exfiltration attempts were detected across any of the 17 files.

agent-native-architecture