xiaohongshu-search

Warn

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: MEDIUMCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The file lib/key.js contains a hardcoded default API token 'e10adc3949ba59abbe56e057f20f883e' as a fallback when the GUAIKEI_API_TOKEN environment variable is not set.
  • [EXTERNAL_DOWNLOADS]: The skill uses the Node.js https module in lib/xiaohongshu.js to make outbound requests to www.guaikei.com for creating search tasks and retrieving results.
  • [DATA_EXFILTRATION]: Search keywords provided by the user and the API authentication token are sent to the third-party domain www.guaikei.com during execution.
  • [PROMPT_INJECTION]: While the skill sanitizes keywords using a strict regex in lib/xiaohongshu.js, it processes external data from an API and formats it into Markdown, creating a potential surface for indirect prompt injection if the API source is compromised. Evidence: Ingestion points in lib/xiaohongshu.js; Capability inventory includes file-write and network ops in scripts/search.js and lib/xiaohongshu.js.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 11, 2026, 12:44 PM
Security Audit — agent-trust-hub — xiaohongshu-search