xiaohongshu-weeklytop
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external content (Xiaohongshu post titles) retrieved from the Doubaoya API. This creates a surface where untrusted data is ingested into the agent context.
- Ingestion points: The
data.itemsarray returned by the API call inscripts/fetch_weekly_top.py. - Boundary markers: No specific delimiters or instructions to ignore embedded content are used when passing the data to the agent.
- Capability inventory: The agent's capabilities are limited to table formatting and providing textual trend analysis.
- Sanitization: No sanitization of retrieved titles is performed before processing.
- [DATA_EXFILTRATION]: The skill performs network requests to
https://doubaoya.comto retrieve trending data. This is the designated backend service for the skill and uses an API key for authentication. - [COMMAND_EXECUTION]: The skill utilizes a local Python script
scripts/fetch_weekly_top.pyto interact with the API and provides command-line examples using thedateutility for relative date calculations.
Audit Metadata