An open source SEO blog advisor that turns your website or topic into a ranked list of blog opportunities — powered by real keyword data and AI.
Live demo: ticketbuddy.ai
- Analyze — Paste your site URL or describe your niche
- Research — Fetches real keyword data via DataForSEO
- Suggest — AI picks the best ranking opportunities for your site
- Write — Generates a full SEO-optimized blog post in seconds
- Free to use — Host your own
- No Auth Needed — No auth needed to login, build your own backend with supabase/postgres
- DataForSEO - Very affordable API costs from DataForSEO. Refer for pricing here: Pricing
- Frontend: Next.js 16 (App Router)
- Keyword data: DataForSEO
- AI: Any OpenAI-compatible provider (Kimi, OpenRouter, Azure, OpenAI)
- Styling: Tailwind CSS v4
See SELF_HOSTING.md for the full setup guide.
Quick start:
git clone https://github.com/ticketbuddyai/keywordbuddy.git
cd keywordbuddy
npm install
cp .env.example .env.local
# fill in .env.local
npm run devKeywordBuddy works with any OpenAI-compatible API. Configure via AI_* env vars:
| Provider | Quality | Cost |
|---|---|---|
| Kimi K2-Thinking | Best | Medium |
| DeepSeek R1 (via OpenRouter) | Great | Low |
| GPT-4o / o3-mini | Good | Medium |
The thinking model (AI_THINKING_MODEL) is used for blog suggestions and article writing. A reasoning/thinking model gives significantly better output, but standard models work too — the thinking display in the UI will simply be hidden.
See CONTRIBUTING.md.
See SECURITY.md for how to report vulnerabilities.
MIT — see LICENSE.