Developer Get Started
OptimAI gives builders several ways to add live intelligence to agents and applications.
Choose An Integration Path
| Path | Use when |
|---|---|
| OptimAI Search MCP | You want an MCP-compatible agent or coding tool to call OptimAI Search as a tool. |
| Search API | You want server-side access to source-backed search results. |
| x402 SDK | You want paid, agent-native search flows with HTTP 402-style payment handling. |
| Claw preview APIs | You are designing extraction, monitoring, or workflow jobs that Claw can execute. |
| Persona preview APIs | You need user-approved memory objects for personal or domain-specific agents. |
Current Public Builder Surfaces
@optimai-network/search-mcp: MCP server for OptimAI Search.@optimai-network/x402-sdk: typed client for x402 search flows.- OptimAI Search Skill: installable guidance for AI coding agents using Search MCP and x402.
- OptimAI GitHub: public repositories for docs, CLI, search MCP, x402, cookbooks, and related tooling.
Recommended First Build
Start with Search MCP. It gives an agent access to live OptimAI Search without requiring you to design your own API wrapper.
export OPTIMAI_API_KEY="sk-..."
codex mcp add optimai-search \
--env OPTIMAI_API_KEY="$OPTIMAI_API_KEY" \
-- npx -y @optimai-network/search-mcp
Then restart your MCP-compatible host and confirm the optimai-search tools are available.
Tooling Model
Build Principles
- Show citations and source links in user-facing answers.
- Store request IDs for debugging.
- Treat API examples marked preview as design contracts, not final production references.
- Keep API keys server-side.
- Ask for explicit user approval before saving Persona memory.
- Use async job patterns for long-running search, extraction, or campaign tasks.