project-init

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt asks the user for Postgres connection details and the scaffolding writes config/docker/ale mbic/.env entries and invokes an LLM on repository diffs, which could require the model to include secret DB URLs/passwords verbatim in generated files or outputs.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). High risk: the project scaffold includes a tracked pre-push hook that sends commit diffs to an external AI CLI (claude) with a --dangerously-skip-permissions flag and instructs that agent to stage/amend commits, which enables data exfiltration of repository contents (including secrets) and automated, stealthy modification of commits (backdoor-like behavior).