xhs-sentiment-dashboard
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The script
scripts/run.pyperforms network operations usingurllib.requestto communicate withai-skills.ai. These operations are used to parse Xiaohongshu links, create analysis tasks, and poll for results. This is the intended behavior of the skill and targets the vendor's own infrastructure. - [INDIRECT_PROMPT_INJECTION]: The skill processes analysis results derived from external Xiaohongshu comments, which represents an indirect prompt injection surface.
- Ingestion points: Data enters the agent's context through the API responses fetched in
scripts/run.py(e.g., thelabeledCommentsandaiInsightsfields). - Boundary markers: No explicit boundary markers or instructions to treat external data as untrusted were found in
SKILL.mdor the scripts. - Capability inventory: The skill environment has network access via
urllib.requestto the vendor's domain. - Sanitization: No sanitization or filtering logic for the retrieved social media content is implemented in the provided Python script.
Audit Metadata