crw-best-practices

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes example code that embeds an API key literal (api_key="crw_live_...") which encourages putting secret values directly into generated code/commands, requiring the LLM to handle/output secrets verbatim.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow described in SKILL.md includes runtime fetching/scraping/crawling of arbitrary web pages (outsider-authored free text) via search → scrape/crawl and then using the scraped markdown/snippets as LLM context, creating an indirect prompt-injection risk.