docs: add AI tab with sei-skill and agent tooling

ai/index.mdx

@@ -0,0 +1,37 @@
+---
+title: 'Build on Sei with AI'
+sidebarTitle: 'Overview'
+description: 'AI tools for building on Sei — from knowledge injection that makes your AI assistant Sei-aware, to live blockchain tooling that lets it act on-chain.'
+keywords: ['sei ai', 'ai development', 'sei-skill', 'mcp server', 'ai agents', 'blockchain ai']
+---
+
+AI coding assistants are powerful — but they weren't trained on Sei's specifics. Without context, they'll give you generic Ethereum answers: wrong gas patterns, missing dual-address handling, outdated package names, and assumptions about finality that don't hold on a 400ms chain.
+
+Sei provides two complementary tools to fix this.
+
+## Two tools, two roles
+
+<CardGroup cols={2}>
+  <Card title="sei-skill" icon="brain" href="/ai/sei-skill">
+    **Knowledge injection.** Teaches your AI assistant everything about Sei — architecture, precompiles, wallet patterns, gas behavior, and ecosystem — so every answer it gives is Sei-specific from the start.
+  </Card>
+  <Card title="MCP Server" icon="plug" href="/ai/mcp-server">
+    **Live blockchain access.** Gives your AI assistant tools to read and write on-chain: query balances, send transactions, interact with contracts, and monitor network state in real time.
+  </Card>
+</CardGroup>
+
+Use them together: sei-skill makes your assistant think in Sei, the MCP Server lets it act on Sei.
+
+## Build AI agents on Sei
+
+<CardGroup cols={3}>
+  <Card title="Cambrian Agent Kit" icon="robot" href="/ai/cambrian-agent-kit">
+    SDK for building autonomous AI agents with DeFi integrations — staking, lending, swaps, liquidity — powered by LangChain.
+  </Card>
+  <Card title="Agentic Wallets" icon="wallet" href="/ai/agentic-wallets">
+    Wallet infrastructure designed for AI agents: programmable signing, session keys, and policy controls.
+  </Card>
+  <Card title="x402 Payments" icon="credit-card" href="/ai/x402">
+    HTTP-native micropayments so agents can pay per-inference for LLM APIs, data feeds, and other monetized services.
+  </Card>
+</CardGroup>