wechat-daily-report
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted chat records which are summarized by an AI, creating an attack surface for indirect prompt injection where malicious content in the logs could attempt to influence the AI's summarization logic.
- [PROMPT_INJECTION]: Ingestion points: The file
scripts/analyze_chat.pyreads user-provided chat data from a JSON file for analysis. - [PROMPT_INJECTION]: Boundary markers: The prompt template in
references/ai_prompt.mdlacks delimiters to clearly isolate untrusted chat content from the AI's core instructions. - [PROMPT_INJECTION]: Capability inventory: The skill uses Jinja2 for rendering and Playwright for browser-based image generation in
scripts/generate_report.py. - [PROMPT_INJECTION]: Sanitization: There is no evidence of content sanitization or escaping before data is passed to the AI or interpolated into the HTML template.
- [EXTERNAL_DOWNLOADS]: The README documentation contains multiple references and instructions to download a ZIP package from the developer's GitHub repository.
- [COMMAND_EXECUTION]: The workflow involves running Python scripts and utilizing the Playwright automation tool to generate the final report image.
Audit Metadata