Sparki and MCP Inventory Automation With Human Approval Built In
Sparki is Spark’s in-app AI inventory agent. Spark MCP lets approved assistants such as ChatGPT and Claude use the same governed inventory tools, with previews and approval before operational changes.
Onboard with Sparki in app or your own AI assistant over MCP.
30 second overview
Sparki in app. Your assistant over MCP.
See both paths use the same governed inventory context and stop for your approval before anything changes.
Agentic onboarding
Bring your data.Sparki builds the system.
Tell Sparki what you are moving, or ask your own AI assistant through Spark MCP. Both paths inspect, map, repair, validate, and verify your data instead of handing you a migration project.
Understands multi-file exports
Explains fixes and judgment calls
Requires approval before import
Spreadsheet upload still works. It is one input, not the onboarding strategy.
SPARK / ONBOARDING RUN
GOVERNED
IN-APP AGENT
“Onboard these catalog, sales, and purchasing exports. Build my forecast foundation.”
3 files · catalog, sales, purchasing
AGENT TRACE
01 / 06
Audit the workspace
Sparki checks what already exists and which readiness gates are still missing.
Catalog, locations, and history inspected
DATA READINESS
IN PROGRESS
Inventory12%
Sales history7%
Supply signals0%
Interactive illustration based on Spark's live onboarding and MCP workflows.
Complete capability
Everything you need to run Sparki and MCP inventory automation
Built into the same system that tracks stock, orders, purchasing, warehouses, and the decisions between them.
Agentic Import & Onboarding
Ask Sparki, or your own assistant over MCP, to inspect, map, repair, validate, and verify one or many data sources with approval before import.
Smart Email Processing
AI reads incoming emails and attachments to automatically create orders, update records, and extract key information.
Proactive Inventory Alerts
Get intelligent notifications about low stock, slow-moving items, and reorder suggestions before problems occur.
Custom Reports & Analytics
Build custom reports with drag-and-drop. Schedule automated report delivery to stakeholders.
Report Templates
Customize invoice, packing slip, and label templates to match your brand. Multiple templates per document type.
User Management & Permissions
Create user accounts with role-based access control. Define exactly what each user can see and do.
Workplaces & Stations
Configure workstations for specific tasks like packing, receiving, or cycle counting. Optimize workflows.
Integration & Agent Access
Connect channels and approved external assistants to the same governed Spark operating context.
Model Context Protocol (MCP)
Connect Claude, ChatGPT, and other MCP clients to governed Spark tools for onboarding and ongoing inventory work.
Notification Settings
Configure email and in-app notifications for important events. Stay informed without being overwhelmed.
Document Number Formatting
Customize numbering sequences for orders, invoices, and other documents. Include prefixes, dates, and more.
From signal to action
A decision workflow your whole team can follow
01
Ask the onboarding agent
Use Sparki in app or your own AI assistant over MCP to inspect what exists and bring in the data your team already trusts.
02
See the recommendation
Spark turns sales, stock, and lead times into a clear next action with the reasoning attached.
03
Approve and move
Review, adjust, and approve in one place. Your team keeps control without rebuilding the analysis every week.
Sparki and MCP, explained
Answers before you connect an assistant
Understand the agent, the protocol, and the controls that keep inventory work reviewable.
What is Sparki?+
Sparki is Spark’s in-app AI inventory agent. It can inspect connected data, guide onboarding, explain inventory signals, prepare recommendations, and help operators complete governed workflows inside Spark.
What is Spark MCP?+
Spark MCP is Spark’s implementation of the Model Context Protocol. It gives compatible AI assistants access to approved Spark inventory tools and context through scoped credentials, previews, approval gates, and an audit trail.
Can I use ChatGPT or Claude with Spark?+
Yes. Compatible versions of ChatGPT, Claude, and other MCP clients can connect to Spark MCP. Your assistant uses the same governed Spark tools while your team controls access and approves operational changes.
What can an AI assistant do in Spark?+
An approved assistant can inspect onboarding state, data readiness, inventory, demand, suppliers, price books, and planning context. It can also prepare imports or operational actions for review when the connected tools and permissions allow it.
How does Spark keep AI actions under control?+
Spark uses scoped credentials, role-based permissions, previews, explicit approval steps, and an audit trail. Imports and operational writes can be reviewed before they change workspace data.
Are Sparki and MCP workflows included in Spark pricing?+
Sparki and governed MCP workflows are part of the Spark product experience. Each plan includes an AI usage allotment, with the available operational capabilities determined by the plan and workspace permissions.
See your own inventory plan in Spark
Upload the data you already have and turn it into the next decisions your team needs to make.
