skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
cartodb/agent-skills59 installs

carto-find-spatial-data

Discover and subscribe to external spatial datasets via CARTO Data Observatory and partner catalogs.

How do I install this agent skill?

npx skills add https://github.com/cartodb/agent-skills --skill carto-find-spatial-data
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The skill provides instructions and references for discovering and subscribing to spatial datasets via the CARTO Data Observatory platform. It documents how to use CARTO-specific tools (MCP and CLI) to search, inspect, and manage data subscriptions without any detected security risks.

  • Socketpass

    No alerts

  • Snykwarn

    Risk: MEDIUM · 2 issues

What does this agent skill do?

carto-find-spatial-data

The CARTO Data Observatory (DO) is a catalog of curated external spatial datasets — demographics, points of interest, environmental data, mobility, financial — that you can subscribe to and use in your data warehouse. Instead of the user hunting down a US census shapefile, this skill answers: "what's already curated in the DO that would help here?"

This skill covers discovery and subscription. Once subscribed, the data is queryable like any other table: right away in the CARTO Data Warehouse and BigQuery, and in other warehouses once CARTO delivers it — see carto-query-datawarehouse.

Access-path routing. With the CARTO MCP server attached, route discovery through search_data_observatory (method: list_datasets|list_filters|search_variables|get_dataset|sample) — available even on token-authenticated sessions — and subscription management through manage_data_observatory_subscriptions (method: list|subscribe|unsubscribe), which needs an OAuth session. Fall back to the carto do CLI when the server isn't attached, for scripted use, or to subscribe from a token session. The evaluation guidance below (coverage, licensing, cadence) applies on every path. Detection signals: carto-basics/references/access-paths.md.

When to use this skill

  • The user asks for "demographics", "points of interest", "boundaries", "mobility data", "weather", or any other category of external spatial data.
  • The user has internal data and wants to enrich it with external context (e.g. spatial-join their stores against census tracts).
  • The user is exploring what's available before committing to a paid subscription.

If the user already has the dataset in their warehouse, jump to carto-explore-datawarehouse instead — DO is for sourcing new external data.

Quick reference

MCP (search available on any session; subscribe needs OAuth):

search_data_observatory { method: "list_datasets", keyword: "census", country: "usa" }
search_data_observatory { method: "get_dataset", dataset_id }        # inspect before subscribing
manage_data_observatory_subscriptions { method: "list" }
manage_data_observatory_subscriptions { method: "subscribe", body: { source: dataset_id, filter?, filterColumns? } }

CLI fallback (no server / scripted / token-session subscribe):

carto do search --keyword "census" --country usa --json
carto do get <dataset-id> --json
carto do subscriptions --json
carto do subscribe <dataset-id>       # public datasets only

What's in this skill

TopicReference
Discovery: searching, filtering, evaluating datasetsreferences/data-observatory.md
Subscribing: public vs premium, where the data is usable in each warehouse, unsubscribingreferences/subscribe.md

Always-on guidance

  • Free vs paid: many DO datasets are free (US Census, OpenStreetMap-derived, NaturalEarth). Premium datasets (Spatial.ai, Mastercard, AirSage, etc.) require a CARTO subscription contract — get_dataset / carto do get shows the licensing and pricing tier.
  • Where the data is usable depends on the user's warehouse. In the CARTO Data Warehouse and BigQuery it's ready right after subscribing. In Snowflake, Databricks, Redshift, PostgreSQL or Oracle, an Admin requests a transfer in Settings > Data Observatory, and CARTO's team delivers it. See references/subscribe.md.
  • Don't recommend a DO dataset without inspecting it first (get_dataset / carto do get). Its metadata reveals coverage (US-only, EU-only, global), update cadence (monthly, yearly, real-time), and licensing — all critical to whether it fits the user's question.
  • DO is external; named sources are internal. Named sources (covered in carto-explore-datawarehouse/references/named-sources.md) are the user's saved queries; DO is the curated catalog of third-party data. Don't confuse the two.

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/cartodb/agent-skills/carto-find-spatial-data">View carto-find-spatial-data on skillZs</a>