travel-planner

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs the agent to extract an xsecToken from search results and then call a CLI with xsec_token="token值", which requires the LLM to include secret token values verbatim in generated commands/outputs.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The package exposes a remote install flow that runs a raw shell script from raw.githubusercontent.com (curl | bash) and then fetches a GitHub tarball for execution — downloading and executing remote shell archives is a high-risk distribution vector for malware.

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

• Third-party content exposure detected (high risk: 0.88). SKILL.md mandates runtime web/community collection via mcp__grok-search__web_search and Xiaohongshu mcporter (outsider-authored posts/pages), and the agent uses those findings as free-text inputs to generate the LLM response/content for the HTML.