Skip to main content
Glama
pipeshub-ai

PipesHub MCP Server

Official
by pipeshub-ai
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

pipeshub_chat

Ask a question and receive a cited answer grounded in your organization's indexed knowledge; switch to web search for public information and continue multi-turn conversations with the same context.

Instructions

Ask a question, get an answer grounded in the org's indexed data with citations. It reads a few retrieved passages — never a whole document, never a complete list.

Three questions this tool gets WRONG. Check them first:

  • Structure — "what's under this epic?", "which pages are in this space?", "what links to this ticket?", "what's in this folder?" → pipeshub_get_record_content mode:"navigate". Ranking cannot see how records relate.

  • Exhaustive — "how many X?", "list ALL the Y", "every Z" → mode:"navigate", which reports the group's real total. This tool undercounts and will not say so.

  • One named document — summarize it, extract from it, what does it say about X → pipeshub_search for the recordId, then mode:"content".

Everything else about the org's knowledge belongs here: policies, processes, decisions, history, "what do we know about X", and any question spanning several documents.

Internal search (default, chatMode: "internal_search"): the user's documents, files, knowledge base, company policies — anything in their PipesHub-indexed sources (Drive, Box, Confluence, Slack, Gmail, Jira, the org's KB, ...).

Web search (chatMode: "web_search"): current events or public information unlikely to be in the org's knowledge base.

Both are plain-chat modes. Agent chat — pass an agentId from pipeshub_agents — runs against that agent's own prompt, tools and knowledge; quick is its only mode, requires the agentId, and is sent automatically.

  • "What's our policy on Y?" → pipeshub_chat (internal_search)

  • "What's in the news about Z?" → pipeshub_chat (web_search)

  • "Find / locate the file named X" → pipeshub_search (then pipeshub_download_record if the user wants the bytes).

Conversation lifecycle — one tool, both start and continue:

  • First turn: omit conversationId. The server creates a new conversation; capture conversationId from the response.

  • Follow-up turn: pass the conversationId from the previous response. Server-side context is preserved — do NOT replay earlier messages, and filters is ignored on follow-ups (set once at creation).

Only re-omit conversationId (start a fresh conversation) when the user explicitly asks to start over / clear context.

The response contains the AI's answer plus citations. To download a cited document, take citations[*].recordId and call pipeshub_download_record.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
queryYesThe user's question or message for this turn.
agentIdNoOptional PipesHub agent to converse with — the `agentId` from `pipeshub_agents`. When set, this turn runs against that agent's configuration (prompt, tools, knowledge). On follow-up turns pass the SAME `agentId` together with the `conversationId` returned by the previous call. Omit for a plain (non-agent) conversation. If unsure which agent to use, call `pipeshub_agents` first to see the options.
filtersNoWhich sources the answer may use. Leave out to use all sources. Only works on the FIRST turn; later turns keep the first turn's sources.
chatModeNoResponse strategy. The valid values depend on whether `agentId` is set: - WITHOUT `agentId` (plain chat): `internal_search` — answer from the org's indexed knowledge (default) — or `web_search` — answer from the live web. - WITH `agentId` (agent chat): `quick` is the only supported mode and is sent automatically, so this argument can be omitted.
modelKeyNoModel id to use (from `pipeshub_sources` `llmModels[*].modelKey`). Defaults to the org's default LLM.
modelNameNo
conversationIdNoExisting conversation id to continue. Omit on the FIRST turn; on every subsequent turn pass the `conversationId` returned by the previous call. Server-side message history is preserved — do NOT replay prior messages.
modelFriendlyNameNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv2.4.1
    • changedInput schema / properties / filters / description
      Previous value: -"Source scoping for retrieval. Pass `apps` ids from `pipeshub_sources`. Only meaningful on the FIRST turn (when starting a new conversation)."New value: +"Which sources the answer may use. Leave out to use all sources. Only works on the FIRST turn; later turns keep the first turn's sources."
    • changedInput schema / properties / filters / properties / apps / description
      Previous value: -"Source-scoping ids from `pipeshub_sources` — connector instance and / or knowledge base ids, mixed freely. The legacy org-wide `knowledgeBase_<orgId>` id is still accepted on deployments that predate per-KB sources. Empty / omitted means no app-side restriction."New value: +"Connector ids to use. Get them from `pipeshub_sources`, where `kind` is \"connector\". Collection ids go in `kb`, not here."
    • changedInput schema / properties / filters / properties / kb / description
      Previous value: -"Legacy / unused. Leave empty."New value: +"Collection (knowledge base) ids to use. Get them from `pipeshub_sources`, where `kind` is \"knowledgeBase\"."
  2. First observedv2.3.3

TDQS

A5/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond annotations (readOnlyHint=false, openWorldHint=true, destructiveHint=false), the description discloses key behavioral traits: it only reads retrieved passages, never full documents or complete lists; it undercounts and will not say so; filters are ignored on follow-ups; server-side context is preserved so earlier messages must not be replayed. These are critical limitations not captured by annotations, and they directly inform agent decision-making. No contradictions with annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long but meticulously structured: core purpose first, then a highlighted 'Three questions this tool gets WRONG' section, followed by 'Everything else... belongs here', internal vs web search, conversation lifecycle, and response handling. Each section is front-loaded with the most important caveats, and every sentence provides actionable guidance. No redundancy or fluff.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description covers all aspects needed to use the tool correctly: its scope and limitations, the routing to alternative tools, the distinction between chat modes, the conversation lifecycle, parameter usage, and how to handle the response (capture conversationId, use citations to download documents). No output schema exists, but the description explains the response contains 'answer' plus 'citations'. It is complete for a chat tool with this complexity.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Although schema coverage is 75%, the description adds substantial semantic depth to parameters. For chatMode it explains the valid values depend on agentId, and that 'quick' is sent automatically when agentId is set. For conversationId it clarifies the lifecycle (omit on first turn, pass on follow-ups, when to re-omit). For filters it notes they only work on the first turn. It also explains how to obtain agentId from pipeshub_agents, and modelKey from pipeshub_sources. This goes far beyond the schema descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb-resource pair: 'Ask a question, get an answer grounded in the org's indexed data with citations.' It also explicitly states what it reads ('a few retrieved passages — never a whole document, never a complete list'), which clearly scopes the tool. It further differentiates itself from siblings by naming three categories of questions it handles poorly and directing to the correct alternatives, making the purpose unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides explicit when-to-use and when-not-to-use guidance. It lists three question types (structure, exhaustive, one named document) and directs to specific sibling tools, then enumerates what belongs here (policies, processes, decisions, history). It also distinguishes internal_search vs web_search, explains agent chat modes, and gives a complete conversation lifecycle (first turn vs follow-up, when to restart). No ambiguity remains.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools