Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Major scoring improvement: known spammers (Temu, Lowes, kaballeros) now auto-score 90+, plus suspicious TLDs, free email providers, urgency/money/scam keywords, legit company check to reduce false positives.
What This Skill Does
Manages the junk/spam folder in Microsoft 365 Outlook and Exchange mailboxes. Scores each email 0-100 using structural signals like missing unsubscribe links, suspicious sender domains, and known scam patterns. Operates in review mode by default, prompting confirmation before moving emails, with an automatic batch-clean mode for unattended processing.
It applies a transparent, rule-based scoring model on top of Exchange's native filters, exposing exactly which signals triggered each score so thresholds can be tuned per mailbox.
When to use it
- Clearing hundreds of accumulated junk emails from a flooded Outlook inbox
- Rescuing legitimate newsletters mistakenly flagged as spam
- Batch-cleaning a shared Exchange mailbox using a confidence threshold
- Auditing a junk folder before migrating to a new email provider
- Identifying phishing attempts missed by Exchange's built-in filters
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: analyze my junk folder and move obvious spam to the learning folder
- 1Run analyze.mjs with the configured profile and mailbox flags
- 2Display each email with its suspicious score and the contributing pattern breakdown
- 3Prompt user to confirm which high-scoring emails to act on
- 4Execute move-to-learning.mjs for confirmed spam using the message ID
- 5Execute move-to-inbox.mjs for any false positives the user flags
Junk folder reviewed with high-confidence spam moved to the learning folder and false positives restored to inbox
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.