The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The troubleshooting guide includes a command to install the official Lambda Stack using wget -nv -O- https://lambdalabs.com/install-lambda-stack.sh | sh -. This pattern downloads and executes a script directly from the vendor's official domain to configure the ML environment.
[INDIRECT_PROMPT_INJECTION]: The skill describes workflows for cloning external Git repositories and processing datasets from .jsonl files. While this creates an attack surface for data-driven instructions, it is a primary and expected function for ML development environments. The risk is mitigated by the standard practice of user oversight in these environments.
[COMMAND_EXECUTION]: The documentation provides numerous examples of using curl, ssh, and shell scripts to automate instance management and training jobs. These are standard administrative tasks for cloud infrastructure management.
[DATA_EXFILTRATION]: Instructions include integrating with wandb (Weights & Biases) and tensorboard for logging metrics and uploading model artifacts to external dashboards. This represents standard telemetry for machine learning experiments.

lambda-labs-gpu-cloud