The Agent Skills Directory

NO_CODE (SAFE): The file consists entirely of markdown documentation. No executable scripts, shell commands, or programming logic are present.- PROMPT_INJECTION (SAFE): The content provides guidance on creating effective prompts rather than attempting to bypass safety filters. It explicitly identifies prompt injection as a 'Sharp Edge' (risk) that engineers must defend against.- EXTERNAL_DOWNLOADS (SAFE): No external dependencies, package managers (npm/pip), or remote scripts are referenced.- DATA_EXFILTRATION (SAFE): There are no commands or instructions that access sensitive local files or perform network requests.- COMMAND_EXECUTION (SAFE): No subprocess calls or system-level commands are included in the instructions or examples.

prompt-engineer