soia-pkm-library-weread-sync
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted metadata (titles, introductions, and highlights) from the WeChat Reading API and interpolates it into markdown files within the Obsidian vault. This creates a surface for indirect prompt injection if the external metadata contains malicious instructions.
- Ingestion points: API responses from
https://i.weread.qq.com/api/agent/gatewayprocessed inenrich_book_details.pyandsync_weread_highlights.py. - Boundary markers: The skill uses specific markdown headers such as
## π εηΊΏand## π δΉ¦η±δΏ‘ζ―to structure synced content. - Capability inventory: Performs file system writes to the Obsidian vault and network requests to the WeRead API.
- Sanitization: Implements basic sanitization for titles and filenames to prevent path traversal and ensure valid filesystem names.
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the official
Tencent/WeChatReadingdependency via npx. This download targets a well-known service and trusted organization.
Audit Metadata