au-opendata-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_portalsA | List the Australian open data portals this server can query, with dataset counts. Call this when the user asks what data sources are available. |
| search_datasetsC | Search Australian government datasets by keyword. |
| get_datasetA | Get one dataset's full metadata and its list of resources (files). Each resource reports |
| list_organizationsA | List a portal's publishers with how many datasets each has, largest first. Only publishers with at least one public dataset are counted. The first result is the portal's largest publisher; use q to find a specific one beyond the limit. |
| preview_resourceA | Show a resource's columns, their types, a sample of rows, and the total row count. ALWAYS call this before query_resource: you cannot write correct SQL without first seeing the column names. Costs nothing for resources held in the portal's DataStore; for others it downloads and caches the file, which then makes the follow-up query free. |
| query_resourceA | Run a read-only SQL query over one tabular resource and get the rows back. The SQL dialect is DuckDB. The resource is always exposed as the table |
| validate_abnA | Validate an Australian Business Number against the ATO's official checksum. Use this instead of judging an ABN yourself — the check is a deterministic modulus-89 calculation and guessing gets it wrong. |
| au_financial_yearA | Map a date to the Australian financial year, which runs 1 July to 30 June. Use this whenever a question involves an Australian financial year, rather than assuming a calendar year. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Each tool targets a distinct resource or action: portals vs organizations, search vs get for datasets, and preview vs query for resources. The two utility tools (validate_abn, au_financial_year) sit outside the dataset workflow but have clear, non-overlapping purposes.
Nearly all names follow a verb_noun snake_case pattern (list_portals, search_datasets, get_dataset, preview_resource, query_resource). The two exceptions, validate_abn and au_financial_year, are still readable and consistent in casing.
Eight tools is well-scoped for a read-only open-data query server. Each tool earns its place, covering discovery, inspection, and querying without bloat.
The surface covers the read-only lifecycle well: discover portals/orgs, search and inspect datasets, preview and query resources, plus ABN and financial-year helpers. Minor gaps exist around exporting or downloading raw files, but core workflows are complete.