While vendoring launching-evals into NVIDIA/Model-Optimizer PR #1239, reviewers flagged a couple of remaining NVIDIA-internal references in packages/nemo-evaluator-launcher/.claude/skills/launching-evals/SKILL.md. Thanks for the gitlab-master fix in #920 already — these two are the leftovers.

Still-internal content (around lines 60–63 in the "Key Facts" section):

1. PPP terminology with NVIDIA-only account names

   **PPP** = Slurm account (the `account` field in cluster_config.yaml). When the user says "change PPP to X", update the account value (e.g., `coreai_dlalgo_compeval` → `coreai_dlalgo_llm`).
   

   "PPP" is internal NVIDIA terminology, and coreai_dlalgo_* are NVIDIA's internal account names. The underlying point (a SLURM account stored in cluster_config.yaml) is universal — could be reframed as something like:

   SLURM account: the account field in cluster_config.yaml. When the user asks to change it (some teams call this a "PPP"), update the account value (e.g., <your_account> → <other_account>).

2. Internal lustre path in HF cache example

   HF_HOME=/lustre/fsw/portfolios/coreai/users/<username>/cache/huggingface hf download <model>
   

   /lustre/fsw/portfolios/coreai/users/... is NVIDIA's internal lustre layout. Could swap for a generic placeholder like <your_hf_cache_dir> or ~/.cache/huggingface.

Happy to send a PR if that's useful — let me know your preference.