The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using the nemo CLI to manage synthetic data generation. This includes validating configurations, running previews, and creating datasets locally or on a cluster. These operations are the primary purpose of the skill.
[COMMAND_EXECUTION]: The troubleshooting section includes instructions for the agent to ask the user for permission to retry commands with the sandbox disabled if network errors occur. This request for reduced security is properly gated by explicit user consent.
[EXTERNAL_DOWNLOADS]: The skill facilitates downloading persona datasets via the nemo data-designer personas download command and may suggest installing the nemo package if it is missing from the environment. These resources originate from the vendor's official toolchain.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes user-provided descriptions and potentially untrusted seed data to generate and execute Python scripts.
Ingestion points: User input via $ARGUMENTS in SKILL.md and seed files described in references/seed-datasets.md.
Boundary markers: Explicit delimiters for the user-provided descriptions are absent in the prompt templates.
Capability inventory: The skill executes the generated Python code through the nemo data-designer utility and python subprocesses.
Sanitization: No explicit sanitization of the input data is performed before interpolation into the generation logic, although the skill includes a validation step (nemo data-designer validate).

nemo-data-designer-plugin