East

East is a small-feature lab for personal research and product validation.

The project is designed so each feature can run independently. A feature should collect data, save raw evidence, and produce an output that can be reviewed later.

启动

在 cmd.exe 里： cd /d E:\www\east .venv\Scripts\activate.bat east reddit-requirements "google" --max-posts 1 --max-comments 1 看到命令行前面有： (.venv) 就说明已经进入了。 不想每次手动激活的话，也可以直接用虚拟环境里的 Python 跑： cd /d E:\www\east .venv\Scripts\python.exe -m east.cli reddit-requirements "google" --max-posts 1 --max-comments 1 简单记：日常开发建议先激活 .venv；临时执行可以直接调用 .venv\Scripts\python.exe

Feature 1: Reddit Requirements Research

Goal: enter a topic, collect related Reddit posts and comments, then turn the evidence into a requirements analysis.

Current decision:

• Reddit API / PRAW is paused because the Reddit developer application is hard to approve.
• AI API inside this project is paused.
• Apify is used only for Reddit collection.
• The project outputs an evidence pack and raw JSON.
• I will read that evidence pack in this chat and write the Chinese requirements analysis for you.

Install

Use Python 3.11 or newer.

python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -e .

Copy the env example:

Copy-Item .env.example .env

Fill in:

• APIFY_TOKEN: your Apify API token
• APIFY_REDDIT_ACTOR_ID: keep the default unless you want to switch actors

No OPENAI_API_KEY is needed.

Usage

Generate an evidence pack:

east reddit-requirements "AI note taking app" --max-posts 10 --max-comments 20

The command writes files under a daily folder. File names also include a timestamp so repeated runs do not overwrite each other:

• outputs/YYYY-MM-DD/<topic>-evidence-YYYYMMDD-HHMMSS.md: readable evidence pack for me to summarize
• outputs/YYYY-MM-DD/<topic>-raw-YYYYMMDD-HHMMSS.json: structured raw data for debugging and reuse

You can also generate a simple local draft without any AI API:

east reddit-requirements "AI note taking app" --analysis-mode extractive --max-posts 5 --max-comments 10

Human-in-the-Loop Summary Workflow

1. Run the command above.
2. Tell me the topic and generated file path, or paste the evidence pack content.
3. I will write the Chinese requirements analysis in this chat.
4. If needed, we can save that summary back into outputs/.

Project Structure

src/east/
  analyzers/      # Evidence pack and local draft builders
  collectors/     # External data collectors
  features/       # Feature workflows
  cli.py          # CLI entry point
docs/
  ROADMAP.md      # Product and feature plan
tests/            # Unit tests

References

• Apify Python client quick start
• Apify Reddit Scraper Actor