The Agent Skills Directory

[NO_CODE]: The skill is comprised entirely of Markdown documentation describing the RAG (Retrieval Augmented Generation) process. It does not include any executable scripts (.py, .js, .sh), configuration files, or automated tasks.
[SAFE]: The content is purely instructional and follows industry best practices for grounding AI responses. No malicious patterns such as prompt injection, data exfiltration, or obfuscation were identified.
[PROMPT_INJECTION]: While the skill describes how to construct prompts using retrieved data, the examples provided include defensive instructions (e.g., 'Answer using ONLY the provided context') to ensure model adherence to the provided data.

context-retrieval