Fake Blog Comments Generator

Mock blog comments for design systems and moderation testing

A realistic block of mock comments is the fastest way to design a comment UI before the article is even published. Designers use the output to prototype Disqus, Hyvor Talk, Commento, Cusdis or Giscus templates; engineers load it into Storybook to validate threading and reactions; QA teams replay it against moderation flows. The trick is volume and realism — a single happy "Great post!" tells you nothing about overflow, truncation, nested replies or spam detection.

A credible dataset mirrors observed sentiment distribution on mainstream blogs: roughly 60% positive, 25% neutral, 10% negative, 5% spam. Length must vary too — single-line reactions to multi-paragraph essays — so that your CSS handles edge cases (very long URLs, overflowing emoji strings, blockquotes). Nest replies up to three levels to test thread collapsing and "view more replies" affordances.

Anatomy of a mock comment

• Author name & avatar — combine first + last names from a Faker-style list; use a Gravatar-style fallback for the photo.
• Timestamp variation — mix recent ("2 hours ago", "yesterday") with older ("2 weeks ago", "3 months ago") to exercise your relative-time formatter.
• Markdown body — most modern comment systems accept bold, italic, links, inline code, blockquotes and fenced code blocks. Render all of them.
• Reactions — like / dislike / heart counts, edited-flag, replied-to user mention.

Comment systems compared

• Disqus — most popular, heavy script, ad-supported on the free tier.
• Commento — open source, self-hostable, no ads.
• Hyvor Talk — privacy-focused, paid SaaS.
• Cusdis — lightweight (< 10 kB), great for static sites.
• Giscus — backed by GitHub Discussions, free for open source.
• Native WordPress — still widely used; pair with Akismet or Cleantalk for anti-spam.

Spam, bots and moderation

Real-world spam patterns to seed into the mock: links to crypto sites, "Visit my site for SEO services", random Unicode bidi noise, repeated identical comments from different "users". Since 2024, AI-flooded comments (ChatGPT-generated) are an emerging concern; they tend to share stylistic markers — bullet lists, "great article", three-sentence closing — that Perspective API (Google Jigsaw) and academic toxicity classifiers can flag. For privacy, every comment system requires GDPR / LGPD opt-in cookies before storing the visitor identity.

FAQ

Does realism really matter? Yes — designing against three happy comments hides every overflow, truncation and reply-thread bug your real users will hit.

Should the mock support markdown? Almost always — Disqus, Commento, Giscus and Hyvor Talk all parse markdown, so your component needs to render it.

How do I test spam moderation? Seed a few comments with the patterns above (crypto links, repeated content, ChatGPT-style closings) and confirm your Akismet or Perspective integration flags them.

Is the data safe to commit? Yes — synthetic comments are not personal data under LGPD or GDPR, so you can ship them in Storybook stories or fixtures freely.