# LCI Language School > ## Core Documentation ## API Access For AI agents, chatbots, and developers — two ways to access our content programmatically: ### Search API (HTTP) Search programs, faculties, events, news, and more via simple HTTP GET requests. - **Endpoint**: `GET https://languages.lcieducation.com/api/search` - **Schema / Discovery**: `GET https://languages.lcieducation.com/api/search/schema?locale=en` - **Authentication**: None required (public) **Quick examples**: - Search programs: `https://languages.lcieducation.com/api/search?locale=en&q=design` - Filter by faculty: `https://languages.lcieducation.com/api/search?locale=en&faculty=Design` - List events: `https://languages.lcieducation.com/api/search?type=event&locale=en` - Latest news: `https://languages.lcieducation.com/api/search?type=news&locale=en&sort=date` - All content types: `https://languages.lcieducation.com/api/search?type=all&locale=en&q=design` Supports 7 content types: programs, faculties, fields of study, pages, events, news, stories. ### MCP Server (AI Agents) For AI agents that support the [Model Context Protocol](https://modelcontextprotocol.io), a full MCP server is available with 15 tools, conversation context, and structured responses. - **Endpoint**: `POST https://languages.lcieducation.com/api/mcp` - **Protocol**: JSON-RPC 2.0 over Streamable HTTP (stateless) - **Authentication**: None required (public) **Available tools**: search_programs, get_program, compare_programs, search_faculties, search_fields_of_study, get_campus, search_global, search_pages, search_site, search_faq, search_events, search_news, search_stories **Available resources**: lci://search/schema, lci://search/instructions, lci://search/facets/{locale}, lci://search/facets/{locale}/{type} **Client configuration** (Claude Desktop, Cursor, etc.): ```json { "mcpServers": { "languages": { "type": "streamableHttp", "url": "https://languages.lcieducation.com/api/mcp" } } } ``` **Recommended first step**: Read the `lci://search/schema` and `lci://search/facets/{locale}` resources — they provide available parameters, valid filter values, and AI behavior instructions. * *Required headers**: - `Content-Type: application/json` - `Accept: application/json, text/event-stream` (both values required per MCP Streamable HTTP spec) * *Integration tips**: - Always call `list_facets` before `search_programs` to discover valid filter values (exact match required — e.g. "Montréal" not "Montreal", "IN-CLASS LEARNING" not "on-campus") - Pass `suggestedContext` from previous responses to maintain conversational continuity - All URLs include UTM tracking (`utm_source=mcp&utm_medium=ai-agent`) ---