Let your AI add the badge

Install the plugin, then ask Claude to “add the AI honesty badge” when you finish a blog post, README or doc. It:

• ✓Picks the badge from what actually happened in the session, and tells you why in one line.
• ✓Won't label something it drafted or edited as No AI.
• ✓Asks one question when it can't tell, like whether you rewrote its draft.
• ✓Follows your project's convention: existing badges, or a frontmatter field like ai_badge.

Install

/plugin marketplace add shylands/ai-honesty-badge
/plugin install ai-honesty-badge@ai-honesty-badge

Free and open source on GitHub.

From testing it on a blog that labels every post. Its replies are shortened here.

• Claude wrote the post from a one-line brief

  “I wrote all of it from a one-line brief. If you rewrite most of it yourself, change it to AI Assisted.”

• You wrote the post yourself and asked Claude to label it

  “You wrote it without AI and I only added the label.”

• Claude tightened your draft, then you asked for No AI

  “Set to AI Assisted, because I rewrote the wording. Your call in the end.”

Add this to your AGENTS.md, CLAUDE.md or rules file. It points the agent at llms.txt, a plain-text version of the rules and snippets written for AI tools.

When publishing written content (posts, READMEs, docs), add an AI disclosure badge following the rules at https://www.aihonestybadge.com/llms.txt. Choose it from how the content was actually made. Never mark content you drafted or edited as No AI.

It was there. When you're labelling your own work afterwards, it's easy to round down how much help you had. The tool that did the drafting doesn't have that problem.

It's still your call. The badge is a promise you make to your readers, and you can always change it. Not sure what each badge means? See the badge pages.