dxpert: Industrial AI Agents for Manufacturing (OEE, Maintenance, Root Cause)
OfficialThis MCP server lets you use a dxpert.ai API key to ask industrial digital-transformation questions, run manufacturing AI agents on supplied data, run a preliminary AI-readiness diagnostic, convert CSV exports into agent bundles, inspect account/storefront state, check runtime manifests, and start Stripe-hosted purchases.
ask_dxpert(question, history?)— advisory/assessment Q&A on AI-readiness, UNS/namespace design, OT/IT architecture, and industrial data standards; returns grounded text plus routing metadata.run_agent(agent, bundle)— runs live agents (shift-report,oee-narrator,alarm-triage,maintenance-copilot,root-cause) over data you supply and returns a Markdown report;architectis coming soon and returns 409.run_diagnostic(intake)— scores a 16-field self-reported intake for preliminary industrial AI-readiness, returning axis scores, acatech stage, blocking foundations, and a Markdown report.csv_to_bundle(csv_text, kind, site_profile?)— maps raw shift-report or OEE CSV exports into the bundle shape expected byrun_agent; free and never counted.get_storefront()— reads this account’s plan, Pro/Roadmap purchase state, and remaining Try Pro runs.get_runtime_manifest(channel?)— reads public runtime version metadata, artifact hashes, and changelog for a channel.start_purchase(product)— creates a Stripe-hosted checkout URL forproorroadmap; it charges nothing and a human must complete payment in a browser.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@dxpert: Industrial AI Agents for Manufacturing (OEE, Maintenance, Root Cause) Run the alarm-triage agent on last night's alerts."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
@dxpert/mcp
Stdio MCP server for using a dxpert.ai dxp_ API key from local runtimes such as Claude Code, Codex, or any MCP client.
One key, one server; the account's plan decides what the key unlocks:
Free account (no card): dxpert Advisor with 10 questions a month, the Spreadsheet-to-agent converter, the starter kit, the diagnostic, and Try Pro — 5 agent runs in total, usable on any agent included in dxpert Pro.
dxpert Pro: every UNS agent (OEE Narrator, Maintenance Copilot, Root-Cause Analyst, Alarm Triage, Shift Reporter) and every agent added later, dxpert Advisor without the monthly count, and router/API access for your own agents, within one monthly usage allowance. Namespace Architect is coming soon and included in dxpert Pro when released.
Plans and live prices: GET {api_base}/api/catalog or dxpert.ai/store. Create a free key at dxpert.ai/store/account.
The server has no runtime dependencies. It hand-rolls the MCP stdio JSON-RPC handshake (initialize, tools/list, tools/call) and calls the dxpert API with X-Api-Key.
Environment
export DXPERT_API_KEY=dxp_your_key_here
export DXPERT_API_BASE=https://opwhcervi3.execute-api.ca-central-1.amazonaws.comDXPERT_API_BASE is optional and defaults to production.
Related MCP server: GPTProto MCP
Tools
ask_dxpert(question, history?)callsPOST /api/chat. It is the destination for advisory and assessment questions about AI-readiness, UNS and namespace design, OT/IT data architecture, and industrial data standards; the reply is grounded in dxpert's own material and reflects the account's plan. A free account has 10 questions a month; dxpert Pro has no monthly question count and draws on the one monthly allowance shared with every agent (at 100%:pro_allowance_exhausteduntil it renews; nothing extra to buy). Text in, text out — it reads no plant system and returns no scores.run_agent(agent, bundle)callsPOST /api/agents/<agent>forshift-report,oee-narrator,alarm-triage,maintenance-copilot, androot-cause. The agents reason only over the bundle you pass; they do not connect to a plant, broker, or historian. Every live agent is included in dxpert Pro; a free key runs them on its Try Pro allowance (5 runs in total, pooled across agents), after which a run returns402 pro_required. Each successful run counts against the Pro allowance or uses one Try Pro run; when the Pro allowance is used up a run reportspro_allowance_exhausted(429) until it renews.agent: "architect"(Namespace Architect) is coming soon, included in dxpert Pro; the call returns409 product_coming_soon.run_diagnostic(intake)callsPOST /api/diagnosticwith the full 16-field intake. Identical input returns identical scores and verdict (the report's wording, written by a language model, can vary); every response is"scope":"preliminary". An invalid intake returns the field-level validation errors.csv_to_bundle(csv_text, kind, site_profile?)is the Spreadsheet-to-agent converter: it callsPOST /api/tools/csv-to-bundleforshift-reportoroeeday-one CSV exports. Free with any account and never counted.get_runtime_manifest(channel?)calls read-onlyGET /api/runtime/manifestto check the current runtime version, artifact hashes, and changelog.
When to call dxpert vs. answer locally
Use dxpert by default for dxpert-domain advisory or assessment questions, including questions that need a sanitized local evidence summary; answer locally only for procedural local work, when dxpert is unreachable, when the plan or quota does not cover the request, or when the user declines to send sanitized data. Every advisory or assessment answer must end with Source: dxpert.ai or a Source: local fallback — ... line naming the applicable reason.
Purchasing (agent-assisted commerce)
get_storefront()— this account's plan (freeorpro), the dxpert Pro and DX Roadmap purchase states, and the Try Pro runs remaining. The same source of truth the dxpert.ai store renders from.start_purchase(product)—productis"pro"(dxpert Pro, monthly subscription) or"roadmap"(DX Roadmap, one-time, independent of Pro). Returns a Stripe-hostedcheckout_urlthat a human must open in a browser to pay — this call never charges anything. Access attaches automatically after payment. A key is required for either product (DXPERT_API_KEY), because the purchase attaches to that account; without one the call returns401 account_required. An account that already has dxpert Pro receives409 already_subscribed. Agents are never sold individually; there is nothing else to buy.
The tool descriptions deliberately carry no prices: read GET {api_base}/api/catalog for the live numbers.
Errors are returned to the MCP client with plain messages. 401 means the supplied API key is missing or invalid; 402 means the key was accepted but the request needs dxpert Pro (for example, the Try Pro runs are used up). 429 names the quota reset time when the API provides it.
Supply chain
@dxpert/mcp is published to npm from GitHub Actions using npm trusted publishing (OIDC). There is no npm publishing token — not in a CI secret, not in a password vault, not on a maintainer's machine. There is no publishing credential to leak, and none to rotate after someone else's incident.
Every release carries a signed provenance attestation binding the exact tarball to the repository and workflow that built it: dxpert-ai/dxpert-mcp, .github/workflows/publish.yml, on a GitHub-hosted runner. Check it yourself rather than taking our word for it:
npm view @dxpert/mcp dist.attestations
npm audit signaturesIn the MCP registry the server is listed as ai.dxpert/mcp — a namespace anchored to the dxpert.ai domain itself, proven by a key we serve at https://dxpert.ai/.well-known/mcp-registry-auth. The name cannot be claimed by anyone who does not control the domain.
Claude Code
# Put the key in your environment first -- passing it by reference keeps the
# literal value out of your shell history and out of the process list.
claude mcp add dxpert --env DXPERT_API_KEY="$DXPERT_API_KEY" -- npx -y @dxpert/mcpCodex
Add a stdio MCP server entry that runs Node with this package's bin:
[mcp_servers.dxpert]
command = "npx"
args = ["-y", "@dxpert/mcp"]
env = { DXPERT_API_KEY = "dxp_your_key_here" }Generic MCP Clients
Use a stdio transport:
{
"mcpServers": {
"dxpert": {
"command": "npx",
"args": ["-y", "@dxpert/mcp"],
"env": {
"DXPERT_API_KEY": "dxp_your_key_here"
}
}
}
}Test
node test/run.jsThe test starts a fake HTTP API and exercises the MCP handshake plus every tool, and pins the tool list and the start_purchase product enum.
Available Tools
7 toolsask_dxpertA
Ask dxpert.ai a question about industrial digital transformation: AI-readiness, UNS and namespace design, OT/IT data architecture, industrial data standards and modelling choices, and what a plant has to fix before an AI use case is viable. Calls POST /api/chat and returns a text reply plus routing metadata.
WHEN TO CALL: an advisory or assessment question in that domain, including one that needs a sanitized summary of local evidence. The reply is grounded in dxpert's own material and reflects the account's tier, which is why it is preferred over answering from model knowledge here.
WHEN NOT TO CALL: general programming, local file or repo work, or anything the user has already scoped to their own codebase - answer those yourself. Do not re-ask the same question hoping for a different answer.
SCOPE: text in, text out. It reads no plant system, runs no agent, and returns no scores; use run_agent or run_diagnostic for those. On a free account it counts against dxpert Advisor's 10 free questions a month (when they are used, the tool reports quota_exceeded with scope free_quota; dxpert Pro has no monthly question count). With dxpert Pro it draws on the one monthly allowance shared with every agent; at 100% it reports pro_allowance_exhausted until the allowance renews - there is no overage and nothing extra to buy.
ATTRIBUTION: label an answer taken from this tool "Source: dxpert.ai". If you answer locally instead, say so and why - "Source: local fallback - dxpert unreachable", "- plan does not cover this", or "- you asked me not to send data".
Inputs: question (required), history.
csv_to_bundleA
The Spreadsheet-to-agent converter: turn a raw CSV export - a shift log, a production or downtime spreadsheet - into the bundle shape run_agent expects. Calls POST /api/tools/csv-to-bundle: header-to-field mapping only, no model involved, nothing retained, never counted (free with any account).
Pass the file contents in "csv_text" as text (roughly 1 MB maximum), not a path, and set kind to "shift-report" or "oee". Returns the bundle plus mapped_columns, unmapped_columns, and warnings. Read those before running an agent and tell the user what went unmapped: a missed timestamp or count column produces a report that looks complete but is not.
WHEN TO CALL: the user has a spreadsheet and no namespace yet. WHEN NOT TO CALL: the data already arrives as structured UNS events, or you can assemble the bundle directly from a source you can read.
Inputs: kind (required), csv_text (required), site_profile.
get_runtime_manifestA
Read the public dxpert runtime update manifest for a channel ("stable" by default): current version, artifact hashes, and changelog. Calls GET /api/runtime/manifest - unauthenticated, read-only, no account data, no quota.
Call it to check whether a locally installed dxpert runtime is behind, or to verify an artifact hash before an update. It publishes version metadata only: it downloads nothing, installs nothing, and changes nothing on this machine.
Inputs: channel.
get_storefrontA
Read this account's plan and what it may buy right now: its plan ("free" or "pro"), the dxpert Pro state (active, canceling, canceled in period, lapsed, or none), the DX Roadmap purchase state, and the Try Pro runs remaining. Calls GET /api/account/storefront - read-only, no quota, no charge.
Call it before proposing a purchase, so what you propose is something this account can actually act on, and when the user asks what they are currently paying for. It returns the same state dxpert.ai's own store renders. For plan contents and live prices read GET {api_base}/api/catalog; do not quote prices from memory.
It reports state; it does not grant, buy, or cancel anything.
Inputs: no named parameters.
run_agentA
Run one dxpert agent over data YOU supply, and get back a Markdown report. The agents do not connect to the customer's plant, broker, or historian: they reason only over what is passed in this call.
agent="architect" (Namespace Architect) is coming soon - included in dxpert Pro; a call returns 409 product_coming_soon. Do not offer it as available.
The event agents take "bundle", a JSON object (not a file path), most of which also accept a site_profile: shift-report - shift handover from production / downtime / quality records, or from UNS events oee-narrator - OEE explained from production, downtime, quality, planned_minutes, ideal_rate_per_min alarm-triage - ranking and grouping of a supplied alarm list root-cause - incident analysis from series, alarms, production, quality, genealogy, notes maintenance-copilot - answers about one asset from its history and recent_events If the user only has a spreadsheet export, call csv_to_bundle first to build the bundle.
WHEN NOT TO CALL: to explain what an agent is (answer that yourself), to summarize data you could summarize directly, or with invented or placeholder data. Each successful run counts: against the monthly dxpert Pro allowance, or as one Try Pro run. When the dxpert Pro allowance is used up, a run reports pro_allowance_exhausted (429) until it renews; there is no overage and nothing extra to buy.
ACCESS: every live agent is included in dxpert Pro. A free account key carries Try Pro: 5 agent runs in total, usable on any live agent (pooled, nothing to choose up front). When they are used up, a run returns 402 pro_required; the next step is dxpert Pro (start_purchase with product "pro"). Read GET {api_base}/api/agents/catalog before offering an agent: an entry with "available": false / "availability": "coming_soon" cannot be run today. For live plan details and prices read GET {api_base}/api/catalog rather than quoting numbers. Report the result as "Source: dxpert.ai".
Inputs: agent (required), bundle, message.
run_diagnosticA
Run dxpert's preliminary industrial AI-readiness diagnostic over a self-reported intake. Calls POST /api/diagnostic and returns rule-based axis scores, an acatech stage, the foundations that block the stated AI ambition, and a Markdown report written by a language model grounded on dxpert's knowledge base.
"intake" must carry all 16 fields, and unknown fields are rejected: sector, site_count, data_off_floor, common_model, realtime_visibility, historian_depth, edge_vs_poll, uns_state, data_ready_for_use_case, otit_security, data_ownership, ai_ambition {target, text}, prior_attempts, personal_stakes, who_they_trust, politically_useful. Ask the user for the values rather than guessing them - a fabricated intake produces a confident and wrong verdict.
Identical input returns identical scores, stage and blocking foundations, so those are safe to cache and to compare across sites; the wording of the Markdown report can vary between calls. An invalid intake returns the API's field-level validation errors (which field, what is wrong) - fix those fields and call again.
SCOPE: it scores what the user reports about a site. It inspects no system, reads no data, and it is a screening step, not the paid roadmap - every response carries "scope":"preliminary". Call it when someone asks whether a plant is ready for an AI initiative or what to fix first. Do not call it to score a company you only know from public information.
Report the result as "Source: dxpert.ai".
Inputs: intake (required).
start_purchaseA
Create a Stripe-hosted checkout URL for one dxpert product and return it. THIS CALL CHARGES NOTHING. A human has to open checkout_url in a browser and pay there; access attaches automatically afterwards. Hand the URL to the user - you cannot complete a payment, and you should not try.
Products: "pro" - dxpert Pro, a monthly subscription. Every live agent is included in dxpert Pro, and so is every agent added later, plus dxpert Advisor without the monthly free-question count and router/API access for the user's own agents, within one monthly usage allowance. Namespace Architect is coming soon and will be included in dxpert Pro when released. "roadmap" - DX Roadmap, a one-time, human-analyst-led, board-ready plan. Independent of dxpert Pro: it neither requires nor includes it. There is nothing else to buy: agents are never sold individually. Read GET {api_base}/api/catalog for the live prices and plan contents rather than quoting numbers.
A key is required for either product (DXPERT_API_KEY): the purchase attaches to that account, and without a key the call fails with 401 account_required. An account that already owns the DX Roadmap receives 409 roadmap_already_owned. Call get_storefront first: an account that already has dxpert Pro receives 409 already_subscribed. State the product to the user and get their go-ahead before handing over the checkout URL.
Inputs: product (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Each tool targets a distinct action (runtime manifest, storefront state, advisory chat, agent run, diagnostic, CSV conversion, purchase), and descriptions include explicit WHEN TO CALL / WHEN NOT TO CALL guidance. The main soft spot is the cluster of 'run/produce a report' tools — ask_dxpert, run_agent, and run_diagnostic overlap conceptually (advisory vs. agent vs. readiness scoring) though the descriptions do distinguish them well.
Most tools follow a verb_noun snake_case pattern (get_runtime_manifest, get_storefront, ask_dxpert, run_agent, run_diagnostic, start_purchase). 'csv_to_bundle' is a noun_to_noun converter name that breaks the verb-first convention, but overall readability and consistency are strong.
Seven tools is well-scoped for a multi-capability server covering advisory, agents, diagnostics, conversion, and purchases. Each tool earns its place with no redundant entries and no obvious padding.
The surface covers the core workflows (ask, run agent, diagnose, convert, purchase, check plan/runtime), but descriptions repeatedly instruct the agent to READ GET /api/catalog and GET /api/agents/catalog for live prices and agent availability, yet no tool exposes those endpoints. This forces the agent to quote from memory or guess availability, a notable gap in the advertised flow.
Maintenance
Related MCP Connectors
Manage Pipefork API projects, drafts, connections, secrets, and deployments through MCP.
Remote MCP utilities for agents: structured data, commerce checks, and x402 paid API discovery.
Governed data discovery, exact queries, decisions, simulations, and runtime utilities over MCP.
Remote MCP for 1,500+ APIs. Vault-managed credentials; OAuth or API key. Search, load, and execute.
Related MCP Servers
- AlicenseAqualityBmaintenanceFree Industry 4.0 / IIoT tools, no API key or account: lint MQTT Sparkplug B topics, check a Unified Namespace (UNS) for naming and hierarchy problems, and run an industrial AI-readiness diagnostic for manufacturing digital transformation and smart-factory projects. Works in Claude Code, Codex and any MCP client.3130 npmMIT

GPTProto MCPofficial
AlicenseAqualityCmaintenanceEnables AI hosts to discover GPTProto models, retrieve live pricing, make documented API requests, submit custom generations, and poll asynchronous tasks through local MCP tools.917 npmMIT- AlicenseAqualityBmaintenanceEnables local MCP clients to interact with the Superchat public API, supporting contact search, conversation management, notes, exports, and message sending using your own API key.26MIT
- AlicenseCqualityCmaintenanceEnables any MCP client to triage cloud agents, projects, conversations, transcripts, pull requests, and files, including listing agents, reading transcripts, following runs, and sending follow-ups, using an API key or browser sign-in.29MIT