AgentSkillsCN

school-email-processor

从 Gmail 中抓取并处理学校邮件，提取附件中的内容

SKILL.md
--- frontmatter
name: school-email-processor
description: "Fetch and process school emails from Gmail, extract content from attachments"

School Email Processor

Fetches unprocessed school emails from Gmail, extracts text from PDF attachments, and identifies which child each email relates to.

Invocation

code
school:process-emails [--days N] [--dry-run] [--child NAME]

Options

  • •--days N: Number of days to look back (default: 1)
  • •--dry-run: Preview emails without marking as processed
  • •--child NAME: Filter to specific child (Elodie, Nathaniel, Florence)

Process

  1. •Connect to Gmail API using stored credentials
  2. •Search for emails matching school patterns (configured in ~/clawd/school-config.yaml)
  3. •Filter out already-processed emails (tracked in SQLite database)
  4. •For each email:
    • •Extract body text
    • •Download PDF attachments
    • •Extract text from PDFs using PyPDF2
    • •Identify which child based on sender address
  5. •Pass to action extractor for AI processing
  6. •Mark as processed in database

Configuration

See ~/clawd/school-config.yaml for:

  • •Children and their school email addresses
  • •Email patterns to monitor
  • •Gmail API credentials location

Output

Returns list of processed emails with:

  • •Email ID, subject, from address, date
  • •Body text
  • •Attachment text (from PDFs)
  • •Identified child name

Usage

bash
# Process last 24 hours
cd ~/clawd/projects/school-email-automation
python -m school_automation.orchestrator process

# Dry run for last 7 days
python -m school_automation.orchestrator process --days 7 --dry-run

# Process only Elodie's emails
python -m school_automation.orchestrator process --child Elodie

Requirements

  • •Gmail API credentials at ~/.clawdbot/credentials/gmail_token.json
  • •PyPDF2 installed (pip install PyPDF2)
  • •School config at ~/clawd/school-config.yaml