You're on data.world: The Knowledge Graph Gives Everyone the Context. Data Workers Keeps It True
On data.world, now part of ServiceNow? Keep its knowledge graph as the context your people and AI use. Data Workers does the operations work that keeps that context true, behind approvals.
Your catalog runs on data.world. Every table, column and glossary term is a resource in a knowledge graph, so "Headcount" links to the column that computes it, the steward who owns it and the reports that show it. Explorer draws lineage across that graph, Archie answers questions in the catalog and in Slack, governance automations sync tags to Snowflake, and the hosted data.world MCP server hands it all to your AI clients. And since July 2025, data.world has been part of ServiceNow, which is bringing its catalog and governance into the ServiceNow AI Platform. data.world gives your people and AI the context. Data Workers does the operations work that keeps that context true, behind approvals: it catches the day a table stops matching its glossary term, fixes the cause where it lives, routes the catalog change to the steward and leaves a receipt.
Key takeaways
- •data.world keeps its job. The knowledge graph, glossary, stewards, lineage, Archie and governance automations stay where they are. Data Workers works next to them from day one, over the data.world MCP server or API.
- •The ServiceNow question has a plain answer. The acquisition closed in July 2025, and data.world says "data.world Enterprise continues unchanged". Data Workers builds on it as it is and connects natively to ServiceNow for tickets.
- •Drift is caught in the tables, not at the next review. Data Workers profiles the tables behind each governed term and checks them against the definition data.world holds.
- •Fixes land where each owner works. The dbt change goes to the model owner as a diff. Catalog notes and links go to the data.world steward as proposals.
- •Every change leaves a receipt: trigger, diff, blast radius, checks before and after, approver and undo path, linked from the ServiceNow ticket.
data.world is the knowledge-graph catalog. Data Workers is the control plane.
data.world records what your data means. The tables underneath keep moving, and Data Workers runs that side, measured against what the knowledge graph says.
Here is a night at a national retailer that runs HR on Workday and service management on ServiceNow. Matillion loads Workday into Snowflake, dbt Cloud builds the people models, Tableau shows the Workforce Overview to HR leaders, and data.world holds the glossary. The "Headcount" term reads "active regular and fixed-term employees; excludes contingent and seasonal workers", is linked to fct_headcount.employee_count, and has the People Analytics lead as steward. The CHRO's staff meets Wednesday at 11:00. This is an illustrative example.
| Time | System | What happens |
|---|---|---|
| Tue 16:40 | Workday | HR operations adds a worker type, "Seasonal", for 1,412 holiday staff in the distribution centers. Before, they were hired as contingent workers |
| Wed 01:30 | Matillion | The nightly Workday job loads the new value into raw_workday.workers in Snowflake |
| 02:10 | dbt Cloud | The nightly job builds fct_headcount. Its filter excludes worker_type = 'Contingent', so the seasonal staff count as employees. The not-null and unique tests pass |
| 02:30 | Data Workers | run_quality_check on fct_headcount flags volume up 9.8% overnight against a weekly range under 0.5% |
| 02:33 | data.world | Data Workers reads the Headcount term and its links over the data.world MCP server: the definition excludes seasonal workers, and the term points at employee_count. get_root_cause and trace_cross_platform_lineage follow the extra rows back through Matillion to the new Workday value. blast_radius_analysis shows the Tableau data source behind the Workforce Overview, fct_labor_cost_per_head and the board pack extract |
| 02:38 | Spellbook and Slack | Data Workers drafts one change set: a dbt diff that replaces the exclusion with an explicit list of the worker types the term counts, an accepted_values test so the next new type fails the build instead of the number, and a note plus a link from the term to the new test for the data.world steward. Approval requests reach the analytics engineer and the steward in Slack |
| 02:40 | ServiceNow | Data Workers opens a ticket for the People Analytics group with the finding, the affected reports and a link to the proposal in Spellbook |
| 08:50 | Spellbook and GitHub | The analytics engineer reviews the diff and blast radius, approves and merges it |
| 09:05 | dbt Cloud | The owner reruns the job in dbt Cloud after approval; it succeeds |
| 09:25 | Snowflake | Data Workers verifies: zero seasonal rows in fct_headcount, employee_count back within the weekly range plus Tuesday's net hires. It writes the receipt in Spellbook, behind the link in the ticket |
| 09:40 | data.world | The steward reviews the proposal and applies the note and the link to the test. The steward's assistant reads the term back over data.world's MCP server and Data Workers adds the change to the receipt |
| 10:30 | Tableau | The Workforce Overview extract refreshes on its schedule from the corrected table. The service desk agent resolves the ticket |
data.world did its job: the definition was clear, linked to the right column, with a named steward. Nobody had told the code. The fix lived in a dbt model data.world doesn't run, and the catalog change belonged to the steward. Both went through a named person, and the HR leaders saw the right number at 11:00.
| Job | What data.world does | What Data Workers does |
|---|---|---|
| The meaning | Holds glossary terms, definitions and stewards as resources in a knowledge graph | Reads them as governed context and measures the tables against them |
| The map | Links terms to columns and draws data and business lineage in Explorer | Joins that graph to live dbt runs, quality profiles, usage and Snowflake metadata |
| The drift | Shows what its harvests and stewards have recorded | Catches the night a table stops matching its term and finds the cause |
| The fix | Records the decision and runs governance workflows | Proposes the change where the cause lives, such as a dbt diff, and queues the rerun |
| The catalog change | The steward applies notes, links and status | Drafts them as proposals with the evidence attached |
| The automations | Queues governance automations such as Metadata Completeness and Snowflake tag sync | Checks the outcome in the tables afterwards, where the automation's own call ends at "submitted" |
| The proof | Keeps the record of what each term means | Verifies downstream and writes a receipt linked from the ServiceNow ticket |
Why doesn't data.world just do this itself?
Because data.world built the product people trust to say what data means, and that trust depends on being the neutral record for every system around it. Its write surface follows from that job. The MCP server changes the catalog: it updates descriptions, tags and custom properties, registers resources, clears stale fields and creates or removes links between resources. Every action runs inside two permission layers, the token's scope and the user's own access in data.world, and edits may be unavailable when an organization turns on lockdown mode. Governance automations run asynchronously, and data.world's docs say it plainly: you learn that the job was submitted, not whether it completed or succeeded. That is the right design for a catalog. It keeps the graph authoritative and leaves the data systems to their owners.
Rewriting a dbt model, changing a Snowflake object or rerunning a job is a different product with a different liability: blast radius across every system, scoped credentials in each, a named owner's approval, a rollback path and downstream proof that the fix held. Data Workers is the product on that side of the line, and it hands every data.world change back to the steward who owns it.
Every tool owns a slice. Data Workers covers the whole lifecycle
data.world owns its slice outright: the knowledge-graph catalog that gives people and AI the context. Each point tool adds another console, contract and handoff. Data Workers covers the whole lifecycle with one context, one approval flow and one audit trail, and builds on the meaning you already keep in data.world.
| Stage | Data Workers | data.world | Why we scored it this way |
|---|---|---|---|
| Catalog & Context | 9 | 9.5 | data.world's home stage: a catalog built on a knowledge graph, with a business glossary, data and business lineage in Explorer, SPARQL over the catalog graph and a hosted MCP server for AI clients. Data Workers reads that graph as a first-class source, with provenance, next to runs, tests and usage. |
| Analytics & Insights | 8 | 5 | Workbench and SQL or SPARQL queries over catalog datasets help people explore, and Archie explains and drafts SQL. Data Workers' Insights agent answers from governed context and checks the numbers behind the answer. |
| Data Quality | 8 | 4 | data.world shows data quality badges and can trigger quality automations its customers configure. Data Workers profiles the tables themselves, finds the cause of a failure and proves the fix held. |
| Observability & Incidents | 8.5 | 3 | Lineage in Explorer shows the impact of a change. Data Workers detects, diagnoses, fixes and verifies across systems, with a receipt for every incident. |
| Pipelines & Ingestion | 8.5 | 2 | data.world harvests metadata from pipelines; it does not run them. Data Workers queues reruns through your orchestrator and catches source changes before they spread. |
| Schema & Migration | 8 | 3 | data.world records tables and columns as catalog resources. Data Workers detects schema changes the night they land, assesses their impact and plans migrations in approved waves. |
| Governance & Access | 8.5 | 8 | Governance workflows, stewards, approval statuses, access request approvals and governance automations such as Metadata Completeness. Data Workers runs the request queue, dry-runs grants and keeps the estate matched to the glossary. |
| Security & Privacy | 8 | 5 | Tag application and sync to Snowflake and a sensitive data classification module for cataloged sources. Data Workers flags new sensitive column names in pull request review and proposes masking for the owner. |
| Cost / FinOps | 8 | 2 | data.world does not manage warehouse spend. Data Workers traces Snowflake credits to the dbt model behind them and drafts the fix for its owner. |
| MLOps & Models | 7.5 | 3 | data.world catalogs the data that models and AI use and serves it as context. Data Workers keeps the data under those models healthy and checks it against its definitions. |
How data.world and Data Workers work together
Your people stay where they are: Claude, ChatGPT or Archie for questions, data.world for the glossary and stewardship, ServiceNow for tickets. Spellbook Data Catalog (in preview) is where the data team looks: each finding, proposed change, blast radius, approver and rollback, with approval requests reaching people in Slack or email. Underneath, Data Context Wizard keeps one governed context graph, the Data-Agents Swarm brings 20+ specialist agents, and the Autonomous Data-Conductor runs each fix end to end: detect, diagnose, fix, review, verify, remember.
What Context Wizard does with data.world. It reads the glossary terms, links, stewards and statuses you point it at, with provenance, and joins them to the systems Data Workers reaches natively, including Snowflake, dbt, dbt Cloud, Tableau, Slack and ServiceNow, among 50+ connectors. Agents get one governed view through explain_table, search_across_platforms, trace_cross_platform_lineage and blast_radius_analysis, and check_staleness flags glossary context and business rules nobody has reconfirmed in 30 days. No agent can promote its own work to authoritative; that takes a named person, the way your stewards already work.
Setup over MCP today. Data Workers connects to data.world over its hosted MCP server or API today. The server lives at https://api.data.world/mcp, with nothing to install. OAuth sign-in names the client asking and lasts 30 days; a public API or service-account token also works. Give the account Data Workers uses view access: catalog changes are proposals the steward applies. Data Workers' agents come from the open-source repository: clone it and add start-agent.sh entries to your client config, as the client setup docs show.
// Example: .mcp.json for Claude Code
{
"mcpServers": {
"data-world": {
"type": "http",
"url": "https://api.data.world/mcp"
},
"dw-context-catalog": {
"command": "/path/to/dataworkers-claw-community/start-agent.sh",
"args": ["dw-context-catalog"]
},
"dw-quality": {
"command": "/path/to/dataworkers-claw-community/start-agent.sh",
"args": ["dw-quality"]
},
"dw-incidents": {
"command": "/path/to/dataworkers-claw-community/start-agent.sh",
"args": ["dw-incidents"]
}
}
}List the tools with your client's own command (/mcp in Claude Code). A governance lead can then ask "which certified terms in the People collection no longer match the tables they point at?" and get the definitions from data.world, the lineage and a quality profile of each table in one answer.
Where writes go. The dbt fix goes to the model owner as a diff to merge. Reruns are queued through your orchestrator, or by the owner in dbt Cloud. Ingestion stays with Matillion: Data Workers reads what lands in the warehouse and never triggers a load. Notes, links, descriptions and status changes go to the data.world steward as proposals, and approved facts land in the Context Wizard graph. When a steward triggers a governance automation, Data Workers checks the outcome afterwards and adds it to the receipt. ServiceNow gets a ticket with the finding and the link to the receipt; the service desk agent owns its state.
One request, L0 to L4. The autonomy ladder is set per domain.
- •L0 manual. Connected, not acting. Stewards compare terms and tables by hand.
- •L1 observe. Data Workers profiles the tables behind each governed term, reports every mismatch with lineage and owners and logs what each action would have needed.
- •L2 propose. Data Workers drafts the dbt diff and the data.world proposals with their blast radius. The owner and the steward approve before anything runs.
- •L3 act reversibly. For proven change classes, such as queuing the rerun of an affected model and re-profiling it after a fix merges, Data Workers acts, verifies and records it. Code changes still arrive as diffs, or as pull requests when your team turns on the GitHub pull-request target.
- •L4 autonomous. For a scoped, trusted class in one domain, Data Workers runs the loop end to end and posts the receipt for review. Catalog changes still go to the steward.
Read more on safety, approvals, who owns the agents and where your data goes. The same pattern holds for Collibra and Atlan, and on the ServiceNow side for ServiceNow tickets and ServiceNow AI Agents. For the category, see Data Workers vs a data catalog, bring your own context and our guides to the AI data catalog and the agentic data catalog.
What changes for your team
Catalog teams spend much of their week after the definition is agreed: checking the code still computes it and chasing owners when a number moves. With Data Workers, those jobs run on autopilot at the level you set.
- •Incidents. A source change that bends a governed metric is caught the night it lands, traced, fixed through its owner and recorded.
- •Data quality. A table that drifts from its glossary term gets a cause, a fix in the system that owns it and a downstream check that it held.
- •Cloud spend. Snowflake credits are traced to the dbt model behind them, and the fix is drafted for that model's owner.
- •Access. A grant request arrives dry-run: effective privileges after role inheritance, the sensitive columns it would reach, any policy conflicts and a least-privilege grant with an expiry.
- •Audits. Each fix links the term it serves, the approver, the diff and the rollback path, ready for the steward to attach in data.world.
- •Migrations. Each wave is planned with its parity checks and catalog links, and the owner signs off before it closes.
Stewards get their week back for the decisions only people can make: what a term means, who owns a domain, which assets earn certification.
Keep data.world, or consolidate?
Keep data.world if you love it; Data Workers works with it from day one. Many teams consolidate once Data Workers runs that slice too.
For most teams on data.world, the knowledge graph, glossary, stewardship and Archie belong there. What teams consolidate is the tooling around it: scripts that diff a catalog export against dbt YAML, a separate monitor for the governed tables, spreadsheets that map terms to models. If you are weighing building this layer yourself on the data.world MCP server, read build it ourselves with Claude Code and MCP servers: reading the graph is the easy part; cross-system context, approvals and rollback are the work.
The case for your CFO
The outcome: every number your leaders see matches the definition your catalog team agreed. Board packs, workforce plans and AI answers stop disagreeing.
The risk story is plain. Data Workers reads data.world under its permission model and proposes every catalog change to the steward who owns it. Every change in the estate shows its blast radius, goes to a named approver, lands through your existing dbt review, is verified downstream and leaves a receipt: trigger, diff, approver, checks before and after, and how to undo it. Unanswered requests expire and escalate; they never auto-grant. Autonomy is set per domain from L0 manual to L4 autonomous, and an org-wide stop halts all autonomous dispatch. The agents run in your infrastructure: your data stays in your systems, and the hosted Conductor sees workflow metadata only. Zero migration.
Why now: AI clients read the glossary over MCP today and quote it next to the number, so a mismatch is visible to everyone who asks. The first win is one governed domain, such as people metrics, read-only, with a report of every term whose table no longer matches it. What stays the same: your data.world subscription, stewards, workflows and glossary. Start with a pilot; the path is on the pricing page, and the pilot is credited in full against the first year. See the ROI of agentic data operations.
The sentence to repeat upstairs: "data.world tells everyone what our numbers mean; Data Workers makes sure the numbers keep meaning it, and every fix comes with an approval and a receipt."
Getting started
Start with a pilot. Pick the domain where a wrong number costs the most, connect data.world read-only next to Data Workers, and let it report every mismatch between glossary and tables for a few weeks before turning on the first fix class. The pilot path and plans are on the pricing page, and the pilot is credited in full against the first year.
FAQ
What does the ServiceNow acquisition mean for our data.world catalog? ServiceNow announced the acquisition in May 2025 and closed it in July 2025, saying data.world's catalog and governance will be brought into the ServiceNow AI Platform. data.world retired its free Open Data Community on July 13, 2026, and says "data.world Enterprise continues unchanged". ServiceNow also sells its own ServiceNow Data Catalog, which it describes as working with existing catalogs. Data Workers works with data.world as it runs today, and with ServiceNow natively.
Does Data Workers write to data.world? Data Workers proposes catalog changes, such as notes, links, descriptions or status, and the steward applies them in data.world. The account Data Workers uses can stay view-only. data.world's own MCP server can write, inside its two permission layers, for the people and clients you allow.
How is this different from Archie? Archie helps people find and understand data: it drafts descriptions, enriches resources and answers in the catalog and in Slack. Data Workers works in the data systems underneath: dbt, Snowflake, your orchestrator and your tickets, acting only through approvals and leaving receipts your stewards can review.
Can Data Workers run data.world's governance automations? Your stewards run them, from data.world or their own AI client. data.world's MCP call confirms the job was submitted; Data Workers checks the result in the catalog and the tables afterwards and records it.
Which connections are native, and which go through data.world? Data Workers connects natively to Snowflake, dbt, dbt Cloud, Tableau (read), Slack and ServiceNow, among 50+ connectors. It connects to data.world over the hosted MCP server or API today, and to Workday and Matillion over their APIs.
Where does our data go? The agents run in your infrastructure on every tier and hold the warehouse credentials and model key. Your data stays in your systems, and the hosted Conductor sees workflow metadata only: goals, signals, table names, proposals with diffs, run records and approval handles.
Sources
- •data.world, homepage ("data.world is now part of ServiceNow!"), https://data.world/ (checked Oct 3, 2026)
- •data.world, product overview ("The data catalog platform powered by a knowledge graph architecture"), https://data.world/product/ (checked Oct 3, 2026)
- •data.world, "A fond farewell" (Open Data Community retired July 13, 2026; "data.world Enterprise continues unchanged"), https://data.world/community/goodbye/ (checked Oct 3, 2026)
- •data.world docs, Connect to the data.world MCP server (last modified Sep 16, 2026), https://docs.data.world/en/413221-connect-to-the-data-world-mcp-server.html (checked Oct 3, 2026)
- •data.world docs, Using the MCP connection to interact with your catalog (last modified Sep 3, 2026), https://docs.data.world/en/413222-using-the-mcp-connection-to-interact-with-your-catalog.html (checked Oct 3, 2026)
- •ServiceNow, "ServiceNow Reports Second Quarter 2025 Financial Results" (Jul 23, 2025; data.world announced May 2025, closed July 2025), https://newsroom.servicenow.com/press-releases/details/2025/ServiceNow-Reports-Second-Quarter-2025-Financial-Results/default.aspx (checked Oct 3, 2026)
- •ServiceNow, "ServiceNow launches the real-time data foundation that puts autonomous AI to work across the enterprise" (May 6, 2026; Workflow Data Fabric and ServiceNow Data Catalog available now), https://newsroom.servicenow.com/press-releases/details/2026/ServiceNow-launches-the-real-time-data-foundation-that-puts-autonomous-AI-to-work-across-the-enterprise/default.aspx (checked Oct 3, 2026)
- •Data Workers, Client setup (open-source docs), https://dataworkers.io/opensource-docs/client-setup/ (checked Oct 3, 2026)
- •Data Workers open-source repository, agent tools and start-agent.sh, https://github.com/DataWorkersProject/dataworkers-claw-community (checked Oct 3, 2026)
- •Data Workers, Spellbook Data Catalog product page, https://dataworkers.io/product/spellbook-data-catalog/ (checked Oct 3, 2026)