douyin-works-crawler
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data (Douyin video titles and descriptions) fetched from an external API. This data is used to generate content analysis and highlights.
- Ingestion points: Data enters the system via the
listWorkByAccountAPI call inscripts/douyin_works_fetcher.pyand is stored in theworkslist. - Boundary markers: None identified in the script or output templates to separate untrusted content from system instructions.
- Capability inventory: The skill performs local string analysis (keyword matching for
#,@, and specific terms like '教程') and formats the output into Markdown. - Sanitization: The script does not explicitly sanitize or escape the content fetched from the API before including it in the final report.
- [COMMAND_EXECUTION]: The skill requires the execution of a Python script (
scripts/douyin_works_fetcher.py) to perform its primary function. This script utilizes standard libraries to make network requests to the vendor's official domain (redfox.hk).
Audit Metadata