Strapi 5 Typesense Plugin

Unified Full-Text, Vector, and Hybrid Search Engine for Strapi 5
An open-source alternative to Algolia & Pinecone integration for Strapi CMS.

⚡ Features

• 🔍 Full-Text Keyword Search: Lightning fast indexing & search powered by Typesense.
• 🧠 Vector & Semantic Search: Native vector search support for AI embeddings (float[]).
• 🔀 Hybrid Search: Combine keyword relevance (q) and vector similarity (vector_query) seamlessly.
• 🤖 Auto-Embeddings: Built-in support for Typesense auto-embedding models (e.g. ts/all-MiniLM-L6-v2).
• 🔄 Real-Time Lifecycle Sync: Automatic indexing on entry creation, update, and deletion in Strapi 5.
• 🎨 Admin UI Dashboard: Built with @strapi/design-system for status monitoring and collection mapping.
• 🔐 Secure Proxy Endpoint: Public /api/typesense/search endpoint without exposing admin API keys.

📦 Installation

In your Strapi v5 project:

npm install strapi-plugin-typesense
# or
yarn add strapi-plugin-typesense

⚙️ Configuration

Enable and configure the plugin in config/plugins.js or config/plugins.ts:

module.exports = ({ env }) => ({
  typesense: {
    enabled: true,
    config: {
      apiKey: env("TYPESENSE_API_KEY"),
      nodes: [
        {
          host: env("TYPESENSE_HOST", "localhost"),
          port: env.int("TYPESENSE_PORT", 8108),
          protocol: env("TYPESENSE_PROTOCOL", "http"),
        },
      ],
      collections: {
        "api::article.article": {
          name: "articles",
          facets: ["category"],
          vectorField: {
            name: "embedding",
            numDim: 384,
            autoEmbed: true,
            embedFrom: ["title", "content"],
            modelName: "ts/all-MiniLM-L6-v2",
          },
        },
      },
    },
  },
});

🌐 Public Search API

The plugin exposes a secure content API endpoint at GET /api/typesense/search:

curl -X GET "http://localhost:1337/api/typesense/search?collection=articles&q=strapi"

Hybrid Search Example

curl -X GET "http://localhost:1337/api/typesense/search?collection=articles&q=strapi&vector_query=embedding:([0.12,0.45,...],k:10)"

🚀 Publishing to NPM & Strapi Marketplace

1. Build and Verify Package

npm run build
npm run verify

2. Commit & Push

git add .
git commit -m "feat: initial release of strapi-plugin-typesense for Strapi 5"
git push origin main

3. Publish to NPM

npm login
npm publish --access public

4. Submit to Strapi Marketplace

Visit Strapi Marketplace Submission and submit your repository:

• Package Name: strapi-plugin-typesense
• Repository: https://github.com/cartulina/strapi-plugin-typesense

📄 License

MIT © Juan David Arias