data.gouv.fr MCP
Official MCP server for data.gouv.fr, the French national Open Data platform. Enables AI assistants to search, explore, and analyze public datasets through natural language conversation. Provides tools for dataset search, resource inspection, tabular data querying, dataservice/API discovery, and usage metrics.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
French government open data API (data.gouv.fr). Public data — low security risk. API key for write operations. HTTPS only. Government data quality varies by dataset.
⚡ Reliability
Best When
You need to explore French government open data programmatically or want an AI assistant to help analyze public French datasets without manual browsing.
Avoid When
You need non-French open data, or need to write/upload datasets. Also avoid if you need STDIO or SSE transport — only streamable HTTP is supported.
Use Cases
- • Searching French government open datasets by keyword
- • Querying and filtering CSV/Excel data from public datasets via natural language
- • Discovering external APIs registered on data.gouv.fr with OpenAPI specs
- • Retrieving usage metrics (visits, downloads) for specific datasets
- • Parsing and analyzing large public data files up to 500 MB
Not For
- • Non-French open data (use country-specific equivalents)
- • Private or restricted datasets
- • Writing or uploading data to data.gouv.fr
- • Real-time streaming data
Interface
Authentication
No authentication required. All tools are read-only. Public endpoint freely accessible.
Pricing
MIT licensed. Official government project, hosted instance is free to use.
Agent Metadata
Known Gotchas
- ⚠ Only supports streamable HTTP transport — no STDIO or SSE
- ⚠ Tabular API has size limits: CSV <= 100 MB, XLSX <= 12.5 MB
- ⚠ Most dataset descriptions and metadata are in French
- ⚠ Claude Desktop requires npx mcp-remote proxy for HTTP transport
- ⚠ Demo environment has separate data from production
Alternatives
Full Evaluation Report
Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for data.gouv.fr MCP.
Scores are editorial opinions as of 2026-03-06.