Moderation Template

Content moderation workflow with scanning, classification, and action.

Chain Flow

content-scanner → classifier → flagger → reviewer → actioner

Agents Included

• content-scanner — Scans content for issues
• classifier — Classifies content type/severity
• flagger — Flags problematic content
• reviewer — Human-in-the-loop review
• actioner — Takes appropriate action

Features

• Content scanning — Automatic content analysis
• Classification — Severity categorization
• Flagging — Highlight problematic items
• Action automation — Automated responses

Usage

mcp__brainbrew__template_bump(template: "moderation")

Then restart Claude Code and use:

"Review content for moderation"
"Moderate user submissions"

Flow Config

flow:
  content-scanner:
    routes:
      classifier: "Scan complete"

  classifier:
    routes:
      flagger: "Classified"

  flagger:
    routes:
      reviewer: "Flagged for review"
      actioner: "Auto-action possible"
    decide: |
      If needs REVIEW → "reviewer"
      If AUTO-ACTION possible → "actioner"

  reviewer:
    routes:
      actioner: "Review complete"

  actioner:
    routes:
      END: "Action taken"