The Agent Skills Directory

[COMMAND_EXECUTION] (LOW): The skill provides instructions to execute mlflow and jq CLI tools. While these are required for the skill's primary purpose of trace analysis, they represent a standard command execution surface.
[PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection (Category 8). * Ingestion points: External MLflow trace data retrieved via the mlflow CLI (SKILL.md). * Boundary markers: Absent. There are no instructions to use delimiters or ignore embedded instructions within the trace data. * Capability inventory: Subprocess execution of mlflow and jq, and file system access for temporary storage. * Sanitization: Absent. While jq is used for structural parsing, the natural language content within the spans is not sanitized before analysis.

analyzing-mlflow-trace