Quickstart

Install and run the open-data ontology MCP — ask for data as concepts and entities, get ranked, failover, refreshed answers.

Install

uv sync                  # base install

# For full data source support (akshare, yfinance, edgar, world bank, …)
uv sync --extra data

Set FINDDATA_ROOT (default: the parent finddata/ dir) to locate the fd-* providers, and EDGAR_IDENTITY="your_email@example.com" before using SEC EDGAR data.

Set up the ontology

# 1. create the ontology tables
fd-open-data-mcp migrate

# 2. import the catalogs (akshare, yfinance, cn-gov, cn-report, edgar, …)
fd-open-data-mcp import-catalog
# or one provider:  fd-open-data-mcp import-catalog akshare

# 3. consume indicator_defs as concepts + propose column→concept bindings
fd-open-data-mcp consume-concepts
fd-open-data-mcp propose-bindings

# 4. seed per-source entity identifiers (stocks, countries, …)
fd-open-data-mcp seed-entities

# 5. generate per-concept refresh schedules
fd-open-data-mcp generate-schedules

Read data

fd-open-data-mcp read --concept-id 234 --entity-type stock --entity-id 1 --date 2024-07-26

Run as an MCP server

fd-open-data-mcp serve        # stdio transport

Or use natural language over the AI search tool (ai_search) to go from query → concepts → entities → values in one call.

Remote access

The live server is exposed for agent clients over HTTPS: add the remote MCP URL https://www.finddatatech.cloud/mcp with the org’s bearer token. The domain swap has shipped — the web playground and remote MCP endpoint are both served over HTTPS.