Use in your browser
Validate a public DCAT-AP catalog with pySHACL against the official SEMIC SHACL shapes, down to the property path behind every violation, or compare two catalogs.
Guest includes 50 lifetime app requests. Create a free account for 100 app requests each calendar month. One allowance across every app, in your browser and your agent.One allowance across every app, in your browser and your agent. Manage usage
Guest access is available after setup. Full account linking verified by PowMCP: Claude only.
Connect this app on its own. Add other PowMCP apps whenever your agent needs another job done.
Open the apps workspace
Choose your agent. Each button opens a new tab with only this app's endpoint ready to add.
ClaudeOpen Claude setup Guest access + account linking verified by PowMCP
VS CodeOpen VS Code setup Guest access available · Account linking not yet verified
KiroOpen Kiro setup Guest access available · Account linking not yet verified
LM StudioOpen LM Studio setup Guest access available · Account linking not yet verified
Visual StudioOpen Visual Studio setup (Windows) Guest access available · Account linking not yet verifiedCopy one ready-to-send prompt for any agent that supports remote MCP connections.
For clients that require manual configuration, use this app-only endpoint, its JSON entry, or one terminal command.
ChatGPT: enable Developer mode in Settings → Security and login, then add this MCP endpoint from the Plugins page. Availability depends on your account and workspace policy. PowMCP has not yet verified ChatGPT account linking. OpenAI setup guide
https://powmcp.com/data-catalog-check/mcp {
"mcpServers": {
"powmcp-data-catalog-check": {
"type": "http",
"url": "https://powmcp.com/data-catalog-check/mcp"
}
}
}Terminal agents add this app with one command:
claude mcp add --transport http powmcp-data-catalog-check https://powmcp.com/data-catalog-check/mcpcodex mcp add powmcp-data-catalog-check --url https://powmcp.com/data-catalog-check/mcpgemini mcp add --transport http powmcp-data-catalog-check https://powmcp.com/data-catalog-check/mcpManage, disable, or remove this connection in your agent's own MCP settings. PowMCP does not label an external connection as installed without confirmation from that client.
Proof
Check a hosted DCAT-AP open-data catalog against the official SHACL shapes, with exact shape-and-focus-node evidence
Validate one public DCAT-AP catalog metadata document against the official pinned SEMIC DCAT-AP SHACL shapes with exact shape, focus-node, and severity evidence, or compare two same-profile catalogs before and after a harvest, with a clear structural-conformance-versus-metadata-quality boundary.
Request cost1 completed result
Guests get 50 lifetime app requests. Free accounts get 100 each calendar month.
dcat_catalog_check
Validates one directly hosted public DCAT-AP open-data catalog metadata document (a portal's RDF export in Turtle, RDF/XML, or JSON-LD, fetched over content negotiation and merged across up to 5 hydra:next pages under a ~50 MB / ~3,000,000-triple cap) against the official pinned SEMIC DCAT-AP SHACL shapes, run offline in a no-network sandbox by pySHACL. Auto-detects the DCAT-AP version (2.0.1, the version the live data.europa.eu MQA harvest enforces, or 3.0.1) or takes an explicit profile. Returns conformance pass/fail, severity counts (Violation/Warning/Info), violations most-severe-first grouped by source shape, focus node, and result path with the shape's message, a structural inventory (catalog/dataset/distribution/data-service counts, catalog records, national-extension namespaces, hydra paging), distinguished states (not-a-dcat-document vs a DCAT graph that fails; version detected vs not-detected vs unsupported; property-missing vs property-invalid), and truncation. Use for one-catalog conformance and diagnosis: 'is our catalog valid DCAT-AP', 'will it pass the MQA harvest', 'what conformance problems does it have'. A pass means only that the fetched graph satisfied the pinned shapes at the named version (SHACL checks structure, cardinality, ranges, and controlled-vocabulary URIs, never whether the metadata is accurate, the datasets exist, descriptions are good, or the portal earns a high MQA quality score), and it does not validate an unhosted local file. Downloads and validates a live multi-hundred-kilobyte catalog and can take up to 90 seconds. Tell the user before calling.
dcat_catalog_compare
Runs the identical pinned pySHACL validation over exactly two directly hosted public DCAT-AP catalogs (baseline first, revised second) sequentially under one shared 115-second deadline, then compares counts among the returned violations. Returns per-catalog hashes, detected DCAT-AP versions, pass verdicts and severity counts; a comparability note asserting both resolved to the same DCAT-AP version and profile (flagging the pair not-comparable otherwise); a difference table of returned violation groups with increased (regressions) or decreased (resolved) counts keyed by source shape and result path; capped group counts cannot establish full-catalog frequencies or issues unique to either catalog; and a ranking by conformance state then fewer Violations. Use for 'did this re-export regress our metadata', 'which conformance issues changed between the two catalogs', and portal-over-portal conformance questions: the same portal before and after a change, or two comparable portals under one profile. The ranking orders machine-checkable conformance only (it never means one catalog's metadata is truer, more complete, or higher quality, and it is not an MQA quality-score comparison), and human review remains required for both. Caller-asserted profile sameness is checked against the detected versions, not assumed.
Other apps for the jobs next to Data Catalog Check.
Scope and boundaries for Data Catalog Check.
Fetch the Schleswig-Holstein Turtle catalog, auto-detect DCAT-AP 2.0.1 or 3.0.1, and retain up to 400 SHACL violations; or compare it with Ireland's catalog under the same detected profile.
Start with documentState, versionState, checkState, and conforms. Conforms means the fetched graph satisfied the selected pinned SHACL shapes; it does not establish metadata accuracy, dataset reachability, description quality, or an MQA score. Fix Violation findings first, treat version fallback and truncation as qualified or incomplete evidence, and use comparison differences only when comparable is true.
A pass means the fetched RDF graph satisfies the pinned SEMIC shapes for the named profile. It does not show that metadata is accurate or current, that described datasets exist, that descriptions are useful, or that a portal will receive a particular MQA quality score.