Skip to main content
Reference for the tools exposed by the PipesHub MCP server. These tools let an AI client (Cursor, Claude Code, Gemini CLI, Claude.ai, LibreChat, etc.) talk to PipesHub — search and chat over the org’s indexed documents, look up people and groups, and download files. The server exposes five hand-written tools that cover the common PipesHub workflows. Each one wraps several lower-level API calls so the LLM gets a single, well-shaped response.

​
pipeshub_chat

Default tool for anything PipesHub-related. Ask a question and get a grounded answer with citations from the org’s indexed sources (Drive, Box, Confluence, Slack, Gmail, Jira, the org’s KB, etc.). When to pick it over the others:
  • “What does document say about X?” → pipeshub_chat (NOT search — search only locates files).
  • “Summarize topic / doc.” → pipeshub_chat.
  • “What’s our policy on Y?” → pipeshub_chat.
  • “Find / locate the file named X” → pipeshub_search (then pipeshub_download_record if the user wants the bytes).
Response: the AI’s answer, citations[] (each with recordId), followUpQuestions, plus the conversationId to use for follow-ups. Vector / semantic search across the org’s indexed documents. Use this only when the user wants to locate a document — by name, topic, or a phrase to grep for. Response: hits[] (recordId, recordName, score, snippet, mimeType, webUrl) sorted by score, plus uniqueRecords[] for deduped record-level info.
For “what does the doc say?” use pipeshub_chat instead — it does the retrieval internally and returns a grounded answer with citations.

​
pipeshub_download_record

Stream the binary content of a single record. Use it when the user wants the actual file bytes (download, attach, open). Response: the file content. Content-Type is forwarded from the upstream service. Binary content is base64-encoded; text is inline.

​
pipeshub_directory

Look up people, groups, and teams in PipesHub. One tool with five actions. Actions:
  • whoami — return the authenticated user’s identity (decoded from the bearer JWT). No other args.
  • list_users — paginated list of org users.
  • get_user — full profile for one user (requires userId).
  • list_groups — paginated list of user groups, with userCount.
  • list_my_teams — teams the caller belongs to, with canEdit / canDelete / canManageMembers flags.

​
pipeshub_sources

Discover available chat sources and AI models in one call. Call this once at the start of a session and cache the result — sources and models change infrequently. Response: up to three sections.
  • sources — every connector instance plus the synthetic knowledgeBase_<orgId> entry. Each id is the value to put in pipeshub_chat’s or pipeshub_search’s apps filter.
  • llmModels — chat / generation models. Each modelKey is the value to pass as modelKey to chat/search. Pick isDefault: true unless the user asks for a specific model.
  • embeddingModels — vector embedding models (only when explicitly requested).

​
Quick Decision Guide