Dashboard-verified spam filter docs: protection levels and Penalty model

[styler]

Part 4 of the stack (merge after #96). Findings from a read-only sweep of the live dashboard with Claude in Chrome, verifying the docs against what streamers actually see.

• Protection levels documented: the Spam Filters page's headline feature — the Filters off / Minimum / Medium / Maximum preset slider — wasn't mentioned anywhere in the docs. The Configuration section now leads with it and the steps match the real flow (expand a filter → General settings / Filter settings / Announcement / Probation).
• Penalty model corrected: the docs described the legacy "Timeout length encodes the action (0 = delete only)" mechanism. The current UI has an explicit Penalty dropdown (Timeout user / Ban user / Delete message / Warn and delete message) plus a separate Duration field. The shared-settings table and "What happens on a violation" now describe that.
• Labels matched to the UI: "Exclude" → "Exclude user group" (Regulars / Subscribers / VIPs / No One), "Timeout message" → "Custom message", "Announcement/Probation settings" as shown.

Related dashboard-verified fixes landed on their own branches: filter defaults + the One-Man Spam threshold scale (0–100, not 0–1) on #94, and module-page label alignment ("Max Tickets", contest duration) on #97. Also confirmed in the sweep: no Language filter in the UI (as #94 states), the Modules page shows exactly the 12 documented cards, and all dashboard paths referenced in the docs (Loyalty → Giveaways / Contests / Loyalty settings) are correct.

🤖 Generated with Claude Code

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
✅ Deployment successful! View logs	docs	0a5caac	Jul 18 2026, 03:03 PM