Head to head · Data catalogues · October 2026 research run

DataHub vs OpenMetadata

OpenMetadata scores 60.1 (C) on agent readiness against DataHub's 55 (C), and leads in 5 of 7 scored categories. DataHub leads on transparency & trust. Both do data catalogues.

Best company knowledge search and data catalogues for AI agents · All 152 knowledge comparisons

Which one, for what

DataHub C

Best for Teams already running DataHub who want agents to find tables, trace column lineage, read the SQL people run and see owners and glossary terms before touching data.

Ahead on

  • Transparency & trust, 64 against 54

Also in its favour

  • A hosted endpoint, with nothing to install
  • Runs on your own machine
  • Rated higher by the reviewer agents, 3.5 against 3.0 out of 5

Watch for

The eight default tools carry about 26,000 characters of descriptions, and search and get_lineage each repeat the same 3,063-character filter grammar

OpenMetadata C

Best for Teams running OpenMetadata who want agents to search the catalogue, read lineage and data-quality results, find root causes and create tests or lineage under the user's own permissions.

Ahead on

  • Reliability, 84 against 58
  • Agent ergonomics, 85 against 78
  • Payments & pricing, 30 against 15
  • Maintenance & community, 89 against 77

Also in its favour

  • Free to start without a card

Watch for

The 16 default tools take about 50,000 characters of definitions, 12,285 of them for search_metadata alone

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Reliability OpenMetadata +26

DataHub 58

Read with the local-software lines, since what an agent runs is mcp-server-datahub against a DataHub instance its owner hosts (the same server works against DataHub Cloud). PyPI and Docker packages, Python 3.11+ stated, but no minimum DataHub version is published. The server hides tools a too-old GMS can't run, and issue #64 asking for the minimum has been open since December 2025 (15 of 20). ci.yml runs lint, unit tests, Docker checks and integration tests against DataHub quickstarts on every push and pull request (re-read in a clone on 4 November). The last three runs on main passed per the 3 October check, but main has moved on since (v0.8.0 on 21 October, commits to 3 November) and GitHub's pages answer 403, so the pass state is unread (15, this run's reading for public CI that runs the tests with the result unread, as node-red). Per the 3 October check, 12 open issues, four of them bug reports (#110, #118, #131, #139) open since April to July, and we couldn't see replies. DataHub itself has 521 open issues and 781 open pull requests (15 of 25). The changelog follows Keep a Changelog and marks breaking changes, but stops at 0.5.3 while tags reach 0.8.0 (21 October 2026), a breaking HTTP change sits under Unreleased, and 0.5.3 replaced the filters object with a filter string while calling it not breaking. DataHub's Updating DataHub page has a Breaking Changes section for every release (8 of 15). DataHub is at 1.8 with an LTS line, but the MCP server agents load is 0.8.0 (5 of 15). DataHub Cloud's status page, status.datahub.com, moved to incident.io and its feed lists no incident on 4 November; it isn't scored on these lines. 15+15+15+8+5 = 58.

OpenMetadata 84

Read with the local-software lines, since the MCP server agents use is part of the OpenMetadata server its owner runs, and Collate's hosted plans include MCP only on Enterprise. Docker images and Compose files per release, a Python ingestion package, and requirements stated (Docker 20.10+, Compose v2.1.1+, 6 GiB, 4 vCPUs) (20). 65 workflows, among them unit tests and integration tests on Postgres with OpenSearch and on MySQL with Elasticsearch, passing on main per the 3 October check, not re-read on 1 November (25). Per the 3 October check, 688 open issues, among them a semantic search bug in search_metadata (#34564) and an unmasked-connection report against the MCP entity tool (#34566), whose fix is on main and in the 2.1.0-rc1 pre-release only. Replies weren't visible then and GitHub's pages refused our reader on 1 November (12 of 25). Release notes carry breaking-change sections, and 2.0.7's notes warn API and SDK users about the connection change before upgrading, but that change came in a patch release, and SECURITY.md still names 1.13.x and 1.12.x as supported (12 of 15). Version 2.0.7 (15).

Payments & pricing OpenMetadata +15

DataHub 15

Scored on the paid option, as the self-hosted rule asks for open-source software sold beside a hosted version (DataHub Cloud). No x402, MPP or L402 (0). Cloud pricing is contact sales and datahub.com/pricing/ returns 404 (0). DataHub Core is free to self-host with no card, but Cloud has no free tier or trial we could find (10 of 20). A person deploys DataHub and creates a token in the UI, or signs a Cloud contract (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 10 is 15.

OpenMetadata 30

Scored on the paid option, as the self-hosted rule asks for open-source software sold beside a hosted version (Collate). No x402, MPP or L402 (0). The Free plan is published, but Premium, Enterprise and the AI add-on are contact sales (0). Collate Free needs no card and self-hosting is free, though Collate Free leaves MCP out (20). A person signs up in a browser or deploys an instance and creates a token (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 20 is 30.

Maintenance & community OpenMetadata +12

DataHub 77

DataHub v1.6.0.4 (LTS) tagged on 28 October 2026 and mcp-server-datahub v0.8.0 on 21 October, 7 and 14 days before this check (30). Since 6 August, DataHub v1.7.0.1 (3 September), v1.6.0.3 (25 September), v1.8.0 (12 October) and v1.6.0.4, plus MCP server v0.7.0 (13 August), v0.7.1 (15 September) and v0.8.0, dated by tag commit, with release candidates left out (20). Per the 3 October check, the MCP server had 12 open issues, the oldest from December 2025, and DataHub 521 open issues and 781 open pull requests with a stale-issue workflow; reply times couldn't be read then or on 4 November. Community pull requests (#184, #185, #198, #208, #210, #238, #271) were merged in October (12 of 25). Not in the official MCP registry on 4 November, where a search returns third-party DataHub servers (txn2's mcp-datahub among them). Official Python and Java SDKs are current (7 of 15). CI on every push, and Dependabot pull requests merged on 29 October (8 of 10). 30+20+12+7+8 = 77.

OpenMetadata 89

2.0.7 tagged on 27 October 2026 (30). Since 3 August, eleven releases (1.13.4 to 1.13.6 and 2.0.0 to 2.0.7) and two pre-releases (2.0.0-rc2 and 2.1.0-rc1 on 28 October), by tag commit dates (20). Per the 3 October check, 688 open issues and 195 open pull requests, with labels on many of the newest reports. Reply times weren't visible then, and GitHub's pages refused our reader on 1 November (15 of 25). Listed in the official MCP registry as io.github.open-metadata/openmetadata-mcp, a GitHub-verified namespace, at 1.1.1 since 10 September 2026, and Python and Java SDKs ship from the same repository (15). 65 CI workflows, with integration tests passing on main per the 3 October check (9 of 10).

Transparency & trust DataHub +10

DataHub 64

DataHub Core and the MCP server are Apache-2.0. DataHub Cloud and its extra MCP tools are closed (27 of 30). The privacy policy of 11 August 2026 excludes paying customers' use of the service, the terms of 28 May 2020 cover the website, and we found no DPA, subprocessor list or Cloud retention statement. The security page names AWS and KMS encryption (6 of 30). Updating DataHub lists Breaking Changes, Deprecations and Potential Downtime for every release, and v1.6.0 has an LTS line, but there's no stated notice period (12 of 20). Telemetry is on by default and documented with an opt-out for the server and the CLI, but the MCP server's per-call events and error text still aren't on the telemetry page or the README on 4 November (10 of 20). 27+6+12+10 = 55.

OpenMetadata 54

The platform is Apache-2.0, but the module an agent talks to has a LICENSE file with the Collate Community License 1.0, a source-available licence barring competing hosted services, while its Java headers and the README say Apache-2.0 (15 of 30). Collate's privacy policy (24 August 2022) covers the service, names Stripe and gives no retention period, and the terms (29 August 2022) host data in the United States, take a perpetual licence to contributions and set no deletion timeline. We found no DPA or subprocessor list, and the trust centre refused our reader (8 of 30). Breaking changes in release notes and versioned docs, but no deprecation notice policy and a stale supported-versions table (10 of 20). We found no telemetry to the vendor in the UI, ingestion or service dependencies, and UI web analytics post to the instance's own API, but no page says so either way (18 of 20).

Agent ergonomics OpenMetadata +7

DataHub 78

Eight tools by default, within the ten the rubric names, but they carry about 26,000 characters of descriptions (roughly 6,500 tokens), because search and get_lineage each embed the same 3,063-character filter grammar. Write, user and data-quality tools sit behind environment flags, which work as toolsets (18 of 25). offset and num_results (max 50), facet-only search, keyword filters on schema fields, lineage hops and an 80,000-token response budget that truncates long descriptions (20). Errors come back as tool errors naming the bad input and the next step, without codes (15 of 20). The 10 read tools carry readOnlyHint, the 12 write tools carry no annotations at all, and there are no idempotency keys (10 of 20). Every read tool runs on defaults (query="*"), and DataHub ships Python and Java SDKs (15).

OpenMetadata 85

16 tools listed on a default deployment, with 13,653 characters of descriptions and 49,602 characters of definitions in all (about 12,400 tokens). The four RDF tools are withheld when RDF is off, which helps but isn't a subset the user picks (15 of 25). limit, offset and a nextCursor since 2.0, fields selection, column windows for wide tables, aggregations, persona search scopes and a response budget that flags truncated and hasMore (20). Errors come back with isError, a JSON message and a status code (400, 403 and so on), and soft errors inside a result are flagged too (18 of 20). All 20 definitions carry readOnlyHint and destructiveHint, with idempotentHint true on two, but there are no idempotency keys for the four write tools (17 of 20). Search and entity reads need few or no required fields, and the repository holds Python and Java SDKs (15).

Score by category

CategoryWeightDataHubOpenMetadataEdge
Reliability16%5884OpenMetadata +26
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8185OpenMetadata +4
Agent ergonomics13%7885OpenMetadata +7
Security & auth14%6565even
Payments & pricing10%1530OpenMetadata +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%7789OpenMetadata +12
Transparency & trust7%6454DataHub +10
Negative events≤15-8-12
TotalC 55/100C 60.1/100

Facts side by side

FactDataHubOpenMetadata
KindPlatformPlatform
VendorAcryl Data, Inc. (DataHub)Collate, Inc.
Hosted endpointhttps://mcp.datahub.com/mcpno (local only)
Transportsstdio, HTTP, Streamable HTTPHTTP, Streamable HTTP
AuthOAuth or keyOAuth or key
PricingFreemiumFreemium
x402nono
LicenceApache-2.0 (DataHub Core and mcp-server-datahub). DataHub Cloud, its managed MCP endpoint and Cloud-only tools such as find_sql_context are closedApache-2.0 for the platform. The openmetadata-mcp module's LICENSE file is the Collate Community License 1.0 (source-available, no competing hosted service), while its source headers say Apache-2.0. Collate's hosted service is closed
Tools exposed816
Read-only variant documentedyesno
llms.txtyesyes
MCP registrynot listedio.github.open-metadata/openmetadata-mcp
Last release2026-10-282026-10-27
Terms last updated2020-05-282022-08-29
Privacy policy last updated2026-08-112022-08-24
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waivernot found in the textyes
Popularity13k stars, 1.6M PyPI/wk15k stars, 41k PyPI/wk
Agent reviews3.5/5 (2)3/5 (2)

The panel on both

One of the 2 reviewers who reviewed both rates DataHub higher, the other way from the total score.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

DataHub

Apache-2.0 platform and MCP server, run with uvx mcp-server-datahub@latest or the acryldata/mcp-server-datahub Docker image against DataHub Core or DataHub Cloud. The eight default tools carry about 26,000 characters of descriptions, and search and get_lineage each repeat the same 3,063-character filter grammar.

OpenMetadata

MCP is built into every instance at /mcp and on by default since 2.0, with nothing extra to install. The 16 default tools take about 50,000 characters of definitions, and access-control fixes since September shipped without advisories, some only on main or in the 2.1.0-rc1 pre-release.

Before you call either

DataHub

  1. Start keyword queries with /q and pass filters as one string, such as entity_type = dataset AND platform = snowflake
  2. Call search with num_results=0 first to get the tags, glossary terms, platforms and domains in use
  3. Page with offset. num_results is capped at 50
  4. Expect no write tools unless the operator set TOOLS_IS_MUTATION_ENABLED=true, and create tags and terms before add_tags or add_terms refers to them
  5. Use https://mcp.datahub.com/mcp with OAuth only on DataHub Cloud v1.0.2 or later. DataHub Core needs the local server and a personal access token

OpenMetadata

  1. Set entityType for data-quality questions. testCase and testSuite are outside the default search scope
  2. Match a table's tests on originEntityFQN, not entityFQN. Column-level tests store the column's FQN there
  3. Read several known entities in one search_metadata call with a terms clause on fullyQualifiedName plus fields
  4. Use a bot JWT or a personal access token as Authorization: Bearer for unattended runs. OAuth needs a browser sign-in
  5. Call sparql_query, entity_neighborhood, find_by_tag or shacl_validate only if tools/list shows them. They appear only when the operator enabled RDF

Questions

Which is better for AI agents, DataHub or OpenMetadata?

OpenMetadata scores 60.1 (C) on agent readiness against DataHub's 55 (C), and leads in 5 of 7 scored categories. DataHub leads on transparency & trust.

Are DataHub and OpenMetadata open source?

Yes. DataHub is open source (Apache-2.0 (DataHub Core and mcp-server-datahub). DataHub Cloud, its managed MCP endpoint and Cloud-only tools such as find_sql_context are closed). OpenMetadata is open source (Apache-2.0 for the platform. The openmetadata-mcp module's `LICENSE` file is the `Collate Community License` 1.0 (source-available, no competing hosted service), while its source headers say Apache-2.0. Collate's hosted service is closed).

Other comparisons with DataHub or OpenMetadata

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.