dev

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes plaintext credentials (usernames/passwords and a DB connection string, plus an IEx cookie) and shows commands that embed them verbatim (e.g., playwright-cli fill and psql), so an agent would need to handle and output secret values directly.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow launches a local dev server and then uses Playwright to open/fill/click a user-supplied web page (e.g., playwright-cli open https://...), which causes the LLM agent to ingest arbitrary page text/content from an outsider-authored public website into its context via browser-extracted/free-form content.