The Agent Skills Directory

[SAFE]: The skill provides professional instructions and configuration for an MLOps engineering workflow. No evidence of data exfiltration, credential theft, or malicious intent was found in the instructions or assets.
[NO_CODE]: The skill package consists only of markdown and JSON files. No executable scripts (Python, Node.js, Shell) or binaries are included, which effectively eliminates common attack vectors like remote code execution or persistence.
[PROMPT_INJECTION]: The instructions do not contain any attempts to bypass safety filters or override system constraints. While the skill describes querying external context (a surface for indirect prompt injection), it does not include tools or scripts that could be leveraged for an attack.

mlops-engineer