run-experiment

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the agent to read the W&B API key from CLAUDE.md and run a command embedding it verbatim (ssh "wandb login <WANDB_API_KEY>") — and also documents setting API keys (e.g., vastai set api-key YOUR_KEY) — requiring the LLM to output secret values directly.