secretary  by zkd8907

AI tool for social media content analysis and WeChat delivery

created 3 months ago
332 stars

Top 83.7% on sourcepulse

GitHubView on GitHub
Project Summary

Secretary is an AI-powered tool designed for automated social media content analysis, targeting users who need to monitor and extract insights from platforms like Truth Social and Twitter. It delivers customized analysis on topics such as finance, politics, and technology directly via WeChat, enterprise WeChat, or QQ.

How It Works

The tool scrapes content from specified social media accounts, processes it using configurable AI prompts for analysis and translation, and then pushes the results through various messaging channels. Its core advantage lies in its flexible prompt engineering, allowing users to define custom analysis dimensions and output formats, and its multi-channel notification system.

Quick Start & Requirements

  • Install dependencies using uv sync or pip install -e . after cloning the repository.
  • Requires configuration of environment variables for Redis, social media access (Truth Social, Twitter), LLM API, and desired notification channels (WeChat, enterprise WeChat, QQ).
  • A config/social-networks.yml file is needed to specify accounts and analysis prompts.
  • Recommended Python version: 3.11+.

Highlighted Details

  • Supports Truth Social and Twitter monitoring.
  • Flexible AI analysis via customizable prompts and support for OpenAI-compatible LLMs.
  • Multiple notification channels: enterprise WeChat (multiple bots), personal WeChat (via Gewechat), and QQ groups.
  • Analysis results are structured in Markdown, including original text, translation, and detailed analytical briefings.
  • Handles LLM JSON parsing errors with retries.

Maintenance & Community

  • The project is maintained by zkd8907.
  • No specific community channels (Discord/Slack) or roadmap links are provided in the README.

Licensing & Compatibility

  • MIT License.
  • Permissive for commercial use and integration with closed-source applications.

Limitations & Caveats

The README mentions that Twitter login generates a session file that is ignored by git. It also notes that if LLM JSON parsing fails after retries, the content processing is skipped.

Health Check
Last commit

6 days ago

Responsiveness

Inactive

Pull Requests (30d)
2
Issues (30d)
0
Star History
47 stars in the last 90 days

Explore Similar Projects

Feedback? Help us improve.