DEV.co
MCP Servers · datagouv

datagouv-mcp

datagouv-mcp is an open-source MCP server that bridges AI chatbots (Claude, ChatGPT, Gemini, etc.) to France's national open data platform (data.gouv.fr). It lets users query datasets conversationally without manual browsing, with a public hosted instance available at no cost.

Source: GitHub — github.com/datagouv/datagouv-mcp
1.6k
GitHub stars
131
Forks
Python
Primary language
MIT
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
Repositorydatagouv/datagouv-mcp
Ownerdatagouv
Primary languagePython
LicenseMIT — OSI-approved
Stars1.6k
Forks131
Open issues12
Latest releasev0.2.29 (2026-06-24)
Last updated2026-06-29
Sourcehttps://github.com/datagouv/datagouv-mcp

What datagouv-mcp is

Python-based Model Context Protocol server exposing data.gouv.fr search and dataset exploration through standardized MCP tools. Supports HTTP transport for multiple AI clients (AnythingLLM, Claude, ChatGPT, Cursor, VS Code, and 10+ others); can be self-hosted or used via public endpoint.

Quickstart

Get the datagouv-mcp source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/datagouv/datagouv-mcp.gitcd datagouv-mcp# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI-Assisted Open Data Discovery

Non-technical users can ask natural-language questions about French open datasets without navigating the web UI directly.

Integration with Enterprise AI Workflows

Teams using Claude, ChatGPT, or open-source LLMs can embed data.gouv.fr exploration into existing AI assistants and chatbots.

Data Analysis & Reporting Acceleration

Analysts and journalists can rapidly locate and retrieve French government and public datasets for analysis.

Implementation considerations

  • Choose between public hosted endpoint (https://mcp.data.gouv.fr/mcp) or self-host; self-hosting requires Python environment and monitoring.
  • Client configuration is MCP-specific; each AI tool (Claude, ChatGPT, Cursor, etc.) has distinct config formats—documentation covers 15+ platforms.
  • No built-in authentication shown in README; verify security posture for production use if handling sensitive user queries.
  • Requires compatible MCP client; older versions of Claude Desktop, ChatGPT, or other tools may not support MCP protocol.
  • Data freshness and schema depend on data.gouv.fr's update cycle; document any SLAs with end users.

When to avoid it — and what to weigh

  • Non-French or Non-French-Language Datasets — The server is purpose-built for data.gouv.fr; if your datasets are hosted elsewhere, use a different MCP server or direct API integration.
  • Offline or Air-Gapped Deployment Required — The public instance requires external network access; self-hosting is possible but adds operational overhead.
  • Real-Time Data Ingestion at Scale — MCP is designed for query/retrieval, not high-volume streaming or ETL pipelines; consider data.gouv.fr's direct APIs for that.
  • Proprietary Data Protection Concerns — All data is from a public platform; if you need to query private datasets, this tool is not suitable.

License & commercial use

MIT License. Permissive open-source license allowing commercial use, modification, and distribution with attribution and no warranty. No compatibility restrictions with proprietary AI platforms identified.

MIT permits commercial use without explicit permission requirement. However, verify that commercial use of connected AI platforms (ChatGPT, Claude, etc.) aligns with their terms—MCP itself does not restrict commercial deployment of this server. No liability or support guarantees included.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationStrong
License clarityClear
Deployment complexityLow
DEV.co fitGood
Assessment confidenceHigh
Security considerations

No explicit security audit, threat model, or OWASP compliance documentation provided in README. Public endpoint at https://mcp.data.gouv.fr/mcp operates without stated rate limiting, authentication, or input validation details—verify production readiness. Self-hosted deployment requires secure HTTP/TLS configuration, network isolation, and monitoring. All queries and data are public (data.gouv.fr), so no PII handling concerns noted; however, user prompts may contain sensitive context depending on client.

Alternatives to consider

Direct data.gouv.fr API (REST/JSON)

If you don't need MCP abstraction, consuming data.gouv.fr's native API directly offers finer control, custom caching, and avoids LLM client coupling.

Other MCP Servers (e.g., for Wikidata, arXiv, or custom data)

If your datasets are not French or not on data.gouv.fr, alternative MCP servers exist for other open-data platforms and knowledge bases.

RAG (Retrieval-Augmented Generation) with LangChain/LlamaIndex

For teams wanting fine-grained vector search, local embedding, and hybrid retrieval, a custom RAG pipeline may offer more control than MCP.

Software development agency

Build on datagouv-mcp with DEV.co software developers

Deploy datagouv-mcp in minutes using the public endpoint or self-host for full control. Choose your AI platform from our 15+ integration guides.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

datagouv-mcp FAQ

Can I use this with ChatGPT for free?
No. ChatGPT MCP support requires a paid plan (Plus, Pro, Team, or Enterprise). Mistral's Le Chat and HuggingChat support MCP on free plans.
Is the public instance at mcp.data.gouv.fr rate-limited or monitored?
README states 'no access restrictions,' but details on rate limits, logging, or monitoring are not disclosed. Contact the maintainers or review terms before high-volume production use.
Can I self-host this behind a firewall?
Yes, but you must expose an HTTP endpoint for MCP clients to reach. Some clients (Claude Desktop) can run local server commands instead of HTTP, simplifying internal deployment.
Does this support non-French languages?
The server supports queries in any language, but datasets are sourced from data.gouv.fr (French open data). Responses and dataset descriptions depend on data.gouv.fr's content language.

Software development & web development with DEV.co

Adopting datagouv-mcp is usually one piece of a larger software development effort. As a software development agency, DEV.co provides software development services and web development expertise — pairing senior software developers and web developers with your team to design, build, and operate mcp servers software in production.

Ready to Connect Your AI to French Open Data?

Deploy datagouv-mcp in minutes using the public endpoint or self-host for full control. Choose your AI platform from our 15+ integration guides.