Content Moderation

Spam + toxic-content scoring using a hybrid of static rules (honeypot, timing, IP reputation), regex/keyword filters, and optional LLM-backed classification for the final 5%. Per-tenant threshold tuning. Audit trail for every moderation decision so customer-support can review borderline cases.

1 project and 2 packages carry this capability.

Projects

Inkwell — Form Submission API
2026-06-09
HTML form ingestion API with spam scoring, multi-destination fan-out (email/Slack/HubSpot/etc), and a 3 KB JS widget.

Packages (2)

@philiprehberger/dotnet-html-sanitizer
dotnet
Whitelist-based HTML sanitizer for XSS prevention with configurable allowed tags, attributes, and URL schemes.
@philiprehberger/rb-sanitize-html
ruby
HTML sanitizer with configurable allow lists, security profiles, and URL/CSS sanitization for safe user content rendering