hexo-blog-manager

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: Orchestrates the execution of bundled Python scripts (generate_cover.py and upload_prep.py) to handle specialized tasks like image generation and Base64 encoding.
  • [COMMAND_EXECUTION]: Uses the GitHub CLI (gh api) to interact with the user's repositories for publishing blog content and hosting static assets.
  • [EXTERNAL_DOWNLOADS]: Connects to the Hugging Face Inference API via the huggingface_hub library to generate custom blog covers based on article titles.
  • [DATA_EXFILTRATION]: Transfers image data and blog markdown to specified GitHub repositories (imHansiy/MyHexo and imHansiy/GitHub_Oss). This behavior is transparent and aligned with the skill's primary function.
  • [PROMPT_INJECTION]: Employs an indirect prompt generation mechanism where blog titles are converted into image generation prompts. The potential for unexpected output is mitigated by a mandatory human-in-the-loop confirmation step before any files are uploaded.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 01:56 PM