Embed makes your wallet, agent, or social app feel more personal with AI-powered recommendations. Deliver real-time, personalized feeds for content, swaps, and bets — directly inside your app, via one simple API. Built for low spam, high signal.
Install the Embed skill to give your AI agent access to on-chain intelligence:
/plugin install embed-ai
Or point any LLM at our machine-readable reference:
curl https://raw.githubusercontent.com/ZKAI-Network/embed-developer/main/llms.txt
| Resource | URL |
|---|---|
| Embed website | getembed.ai |
| Documentation | docs.getembed.ai |
| Console | console.getembed.ai |
| SDK (npm) | @embed-ai/sdk |
| Pipeline SDK (npm) | mbd-studio-sdk |
Serve deployed feeds, manage configs, ingest data. Single endpoint: POST /v3/for-you.
npm install @embed-ai/sdkBuild recommendation pipelines: Search → Features → Scoring → Ranking.
npm install mbd-studio-sdk- 01 — Getting Started — Polymarket recommendations demo: search → features → scoring → ranking pipeline
- 02 — Polymarket Feed — Full-stack Next.js app with personalized Polymarket prediction market feed, real-time charts, and wallet-based recommendations
This repo is also an installable Claude Code plugin:
├── .claude-plugin/
│ └── plugin.json # Plugin manifest
├── skills/
│ └── embed-ai/
│ ├── SKILL.md # What the AI agent learns
│ └── references/ # API docs, examples
├── mcp/ # MCP server (runtime tool access for agents)
├── llms.txt # Machine-readable reference for any LLM
└── examples/ # Working demos
MIT