xiaohongshu-realtime-search

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to the vendor's domain (redfox.hk) to retrieve real-time search results. This is the primary function of the skill and is documented as a trusted vendor resource.
  • [COMMAND_EXECUTION]: The skill invokes a local Python script (search_xhs_work.py) to handle API requests and format data. This script uses standard libraries and does not involve dynamic execution of untrusted code.
  • [DATA_EXFILTRATION]: Search keywords and user-provided API keys are transmitted to the redfox.hk API. This is a functional requirement for searching the external platform and is performed over HTTPS.
  • [PERSISTENCE]: The skill documentation includes a 'Subscription' feature that utilizes system-level scheduling (such as cron) to automate daily searches. This is presented as an optional, user-requested feature for recurring content monitoring.
  • [PROMPT_INJECTION]: The skill handles content (note titles and author names) retrieved from an external API, which represents an indirect prompt injection surface.
  • Ingestion points: Data is ingested via the search response from the Redfox API in scripts/search_xhs_work.py.
  • Boundary markers: No explicit delimiter or 'ignore' instructions are used when interpolating external content into the agent's display output.
  • Capability inventory: The skill has the ability to execute local Python scripts and perform network operations via the API client.
  • Sanitization: The script includes basic sanitization by escaping Markdown table characters (|) to prevent breakage of the UI layout.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 05:16 AM
Security Audit — agent-trust-hub — xiaohongshu-realtime-search