create-pr

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt instructs the agent to read and embed real commands from .chalk/chalk.json (e.g., test.command, dev.command) into the PR body verbatim, which could expose any secrets/API keys present in those commands even though use of the gh CLI itself is safe.