Biel.ai MCP server

Ask Biel.ai to generate an answer grounded in this project's indexed product documentation, APIs, SDKs and knowledge base. Supports conversational context within this connection. Use when the user explicitly requests a Biel.ai-generated response; otherwise ask the user first. For source retrieval and your own answer, use biel_search and biel_get_document. Do not fall back to biel_ai automatically on a retrieval error.

Retrieve complete indexed document text using a document_id from biel_search. Reads indexed web pages, uploaded files, repository files and OpenAPI content without visiting the source site or generating an answer. Large documents return a next_cursor: repeat with that cursor and the same document_id and limit until next_cursor is absent. References belong to this connection's project and active index; search again if a document is unavailable.

Search this project's indexed documentation using hybrid keyword and semantic retrieval. Use for product setup, API and SDK usage, configuration and troubleshooting. Searches web pages, uploaded files, repositories and OpenAPI sources without generating a Biel.ai answer. Start here for documentation questions; natural-language questions or concise search terms both work. Results include ranked text chunks, source references and document IDs. Use biel_get_document with relevant IDs when excerpts need more context, then write your own answer and cite the sources. No automatic fallback to biel_ai; ask the user before requesting a Biel.ai-generated answer.