notion-to-weixin
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires installing the node-wxcli package from the npm registry and uses curl to download cover images from URLs retrieved from Notion page metadata.
- [COMMAND_EXECUTION]: Multiple shell commands are used to perform the workflow, including interaction with notion-cli and node-wxcli, as well as file management and JSON parsing with jq.
- [INDIRECT_PROMPT_INJECTION]: The skill processes Notion page content and utilizes an LLM to generate a summary. This data flow creates a surface for indirect prompt injection. 1. Ingestion points: Notion page content exported to Markdown (page.md) in Step 6. 2. Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the AI summarization step. 3. Capability inventory: The skill can execute shell commands (notion, node-wxcli, curl, jq) and perform file operations. 4. Sanitization: No sanitization or validation of the Notion content is performed prior to summarization.
Audit Metadata