AI 코딩 에이전트를 위한 공식 Twilio Skills and MCP. Claude Code, Cursor, Codex 및 Agent Skills 표준을 지원하는 모든 도구에서 작동합니다.

공개 베타: Twilio MCP 및 Twilio Skills는 현재 공개 베타 제품으로 사용할 수 있으며, 이 저장소에 포함된 정보는 변경될 수 있습니다. 이 기능은 아직 구현되지 않았고 다른 제품이 일반적으로 사용 가능한 것으로 선언되기 전에 변경 될 수 있음을 의미합니다. 공개 베타 제품은 Twilio Support Terms 또는 Twilio Service Level Agreement에 적용되지 않습니다.

---

트위리오 MCP

Twilio MCP 서버는 Twilio 문서 및 API 사양에 대한 검색을 제공합니다. 그것은 두 개의 도구를 노출:

** 엔드포인트:** https://mcp.twilio.com/docs

mcp/README.md]는 모든 지원되는 IDE에 대한 설정 지침을 참조하십시오. 문서: https://www.twilio.com/docs/ai/mcp

피드백

• 질문 및 피드백 : 질문[email protected]
• 문제: GitHub 문제

---

Twilio 기술

Twilio for AI

Official Twilio Skills and MCP for AI coding agents. Works across Claude Code, Cursor, Codex, and any tool supporting the Agent Skills standard.

Public Beta: Twilio MCP and Twilio Skills are currently available as Public Beta products, and the information contained in this repository is subject to change. This means that some features are not yet implemented and others may be changed before the products are declared as Generally Available. Public Beta products are not covered by the Twilio Support Terms or Twilio Service Level Agreement.

---

Twilio MCP

The Twilio MCP server provides search over Twilio documentation and API specifications. It exposes two tools:

Endpoint: https://mcp.twilio.com/docs

See mcp/README.md for setup instructions across all supported IDEs. Documentation: https://www.twilio.com/docs/ai/mcp

Feedback

• Questions and feedback: [email protected]
• Issues: GitHub Issues

---

Twilio Skills

Skills are structured packages of procedural knowledge for AI coding agents. They follow a progressive disclosure architecture: your agent sees lightweight metadata for all available skills, loads the relevant skill when a task matches, and drills into reference material on demand — keeping your context window lean while providing deep domain expertise exactly when needed.

What Skills help with

• Product selection: “I need to verify users” — the agent reasons through Twilio Verify vs. custom OTP, whether to add Lookup for fraud scoring, before any code gets written
• Architecture patterns: “Build me an AI voice agent” — the agent asks questions to understand your use case complexity, then recommends the right combination of products (e.g., Conversation Relay + Memory + Conversation Intelligence + TaskRouter)
• What to avoid: Skills carry explicit CANNOT sections that reduce hallucination by documenting hard constraints and common pitfalls

Available skill categories