zsxq-homework-likes-export
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the ZSXQ API, including solution text and topic titles created by external users, which can serve as a vector for indirect prompt injection.
- Ingestion points: The
fetch_zsxq_task_solutions_signed.mjsscript retrieves solution content and metadata from theapi.zsxq.comendpoint. - Boundary markers: The skill does not implement specific boundary delimiters or safety instructions when processing the external text for CSV generation.
- Capability inventory: The skill has the ability to execute local scripts (Node.js and Python), perform network operations via
fetch, and read/write files in the local environment. - Sanitization: The
build_homework_likes_csv.pyscript applies basic sanitization by cleaning whitespace and truncating solution text excerpts to 180 characters before writing to the CSV output. - [EXTERNAL_DOWNLOADS]: The skill performs network requests to
api.zsxq.comto fetch task and solution data. This is a well-known service in the context of the skill's functionality, and the traffic is restricted to the platform's official API.
Audit Metadata