The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Bash commands such as find, grep, and cat to discover existing prompts and identify the LLM providers used in the local repository. These actions are limited to gathering development context and do not perform any administrative or dangerous system modifications.
[PROMPT_INJECTION]: The instructions provide explicit guidance to the agent on how to write prompts that are resistant to injection, recommending the use of clear delimiters (e.g., XML tags) to separate instructions from untrusted user data.
[SAFE]: No indicators of malicious activity, data exfiltration, or obfuscation were found. The skill's operations are well-documented and align with its stated purpose as an engineering assistant for prompt design.

cortex-prompt