The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns, obfuscation, or injection attempts were detected in the provided markdown file.
[NO_CODE] (INFO): The skill consists entirely of conceptual instructions and architectural patterns for RAG. There are no associated scripts (Python, JS, etc.), configuration files for package managers, or command-line instructions.
[DATA_EXPOSURE] (SAFE): No hardcoded credentials, sensitive file paths, or network exfiltration patterns are present.
[INDIRECT_PROMPT_INJECTION] (LOW): While the skill describes processing external data (RAG), it does not provide an implementation with side-effect capabilities (like file-writing or network requests) that could be exploited by malicious content within processed documents.

rag-implementation