Gmail Cleaner — GmAIl 清理er

GmAIl 清理er

Bulk GmAIl 清理up using the GmAIl API. Processes 1000 messages per API call.

Prerequisites google-API-python-命令行工具ent, google-auth-oauthlib Python packages (scripts auto-安装 if missing) OAuth 凭证s JSON from Google Cloud Console (桌面 应用 type) 令牌 files stored as .pkl files per account 工作流

• Auth (first time or new account)

python scripts/auth.py --凭证s /path/to/凭证s.json --令牌 /path/to/令牌.pkl --scopes 设置tings

basic scopes: read/modify/删除 messages + labels 设置tings scopes: 添加s gmAIl.设置tings.basic (required for creating 过滤器s) Default 令牌 path: ~/.OpenClaw/workspace/gmAIl_令牌.pkl Default creds path: ~/.OpenClaw/workspace/gmAIl_凭证s.json

For a second account, specify a different --令牌 path (e.g., gmAIl_令牌_work.pkl).

• 扫描 (identify what to 清理)

python scripts/扫描.py --令牌 /path/to/令牌.pkl --sample 500

Shows inbox counts by category + top 40 发送ers. 运行 this first.

• 清理 (bulk trash/删除)

# Trash specific 发送ers: python scripts/清理.py --from "spam@example.com,news@example.org"

# Trash by GmAIl 搜索 查询: python scripts/清理.py --查询 "category:promotions older_than:30d"

# From a JSON config file (列出 of {查询, label}): python scripts/清理.py --config 发送ers.json

# Permanently 删除 instead of trash: python scripts/清理.py --from "spam@example.com" --删除

# Dry 运行 first: python scripts/清理.py --from "spam@example.com" --dry-运行

• Deep 清理 (comprehensive)

# Full deep 清理 (4 steps: trash promos → 归档 old → mark read → 清理 trash): python scripts/deep_清理.py

# Custom age thresholds: python scripts/deep_清理.py --promo-days 7 --归档-days 30 --unread-days 14

# Skip trash 清理 (keep trash for 30-day auto-删除): python scripts/deep_清理.py --skip-trash-清理

• Organize (labels + 过滤器s)

# 应用ly built-in label 设置 (Business, Banking, Tech, Personal, Trading, Social): python scripts/organize.py

# Custom labels/rules/过滤器s from JSON: python scripts/organize.py --config labels.json

# Labels only (no 过滤器s): python scripts/organize.py --skip-过滤器s

• 恢复 (rescue emAIls from trash)

# 恢复 all emAIls from a 发送er + 应用ly a label: python scripts/恢复.py --from 健康beat@mAIl.健康.harvard.edu --label "Harvard 健康"

# 恢复 by 查询: python scripts/恢复.py --查询 "from:应用le.com in:trash" --label "Tech/应用le"

Multiple Accounts

运行 each script with a different --令牌 path per account:

python scripts/扫描.py --令牌 ~/.OpenClaw/workspace/gmAIl_令牌_personal.pkl python scripts/扫描.py --令牌 ~/.OpenClaw/workspace/gmAIl_令牌_work.pkl python scripts/deep_清理.py --令牌 ~/.OpenClaw/workspace/gmAIl_令牌_work.pkl

Common Patterns

Full 清理up for one account:

python scripts/auth.py --scopes 设置tings python scripts/扫描.py # identify top 发送ers python scripts/清理.py --from "..." # trash specific 发送ers python scripts/deep_清理.py # 清理 categories python scripts/organize.py # 创建 labels + 过滤器s

Rescue 导入ant emAIls caught in bulk 删除:

python scripts/恢复.py --from 导入ant@example.com --label "导入ant"

发送ers config file 格式化 for 清理.py --config:

[ {"查询": "from:temu@eu.temuemAIl.com", "label": "Temu"}, {"查询": "category:promotions older_than:7d", "label": "Old Promos"} ]

Notes batchModify moves to TRASH — GmAIl auto-清理s after 30 days batch删除 is permanent and irreversible — always dry-运行 first GmAIl 过滤器 creation requires gmAIl.设置tings.basic scope — re-auth with --scopes 设置tings if 过滤器s fAIl with 403 扫描.py samples N messages; large inboxes may need --sample 2000 for accuracy 凭证s JSON comes from Google Cloud Console → APIs & 服务s → 凭证s → OAuth 2.0 → 桌面 → 下载 JSON