Skip to main content
Glama
mcp-french-public-procurement

mcp-french-public-procurement

by Hug0x0

mcp-french-public-procurement

フランスの公共調達データ(DECP、BOAMPの探索、発注者、供給者、SIRENEのコンテキスト)のためのMCPサーバー。

ツール

MCPを実行し、最初にfrench_public_procurement_get_sourcesを呼び出してソースのカバレッジを確認してください。このサーバーは、上記のトピックに関するドメイン固有のツールも公開しています。

Related MCP server: mcp-marches-publics

インストール

npm install
npm run build
npm test
npm run dev

Claude Desktop

{
  "mcpServers": {
    "french-public-procurement": {
      "command": "npx",
      "args": ["mcp-french-public-procurement"]
    }
  }
}

ソース

公開

docs/publishing.mdを参照してください。

Glama / Docker

リポジトリにはDockerfileとglama.jsonが含まれています。

ビルド手順:

["npm install", "npm run build"]

CMD引数:

["node", "dist/index.js"]

安全性

このMCPは、エージェントが公開ソースを発見・要約するのを支援します。公的な権威ではありません。決定事項は、所管の公共サービスまたは元のデータ作成者に照らして確認してください。

ライセンス

MIT

Available Tools

5 tools
french_public_procurement_build_supplier_watchB

Build useful source queries for monitoring one supplier or buyer across DECP, BOAMP and SIRENE context.

Inputs: name (required), siret.

french_public_procurement_fetch_source_excerptC

Fetch a short text excerpt from a curated source by index or title keyword.

Inputs: source_key (required), max_chars.

french_public_procurement_get_sourcesA

List curated sources used by this MCP.

Inputs: no named parameters.

french_public_procurement_search_datasetsC

Search data.gouv.fr for public-procurement datasets such as DECP, BOAMP, concessions, buyers, and PLACE exports.

Inputs: query, page_size.

french_public_procurement_search_decpC

Query the consolidated DECP public-procurement dataset on data.economie.gouv.fr.

Inputs: year, buyer, limit, supplier.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.3/5.0

    Scored across 5 tools

    Disambiguation4/5

    Each tool has a reasonably distinct role: source listing and excerpt fetching differ from the dataset/DECP search tools, and build_supplier_watch creates queries rather than executing them. The only mild ambiguity is between search_decp and search_datasets, but the descriptions clearly distinguish searching a specific dataset from searching available datasets.

    Naming Consistency5/5

    All tools share the same french_public_procurement_ prefix and use a consistent snake_case verb-noun pattern: get_, fetch_, search_, search_, build_. There is no mixing of styles or unpredictable naming.

    Tool Count5/5

    Five tools is well-scoped for a niche French public procurement MCP. The set avoids bloat while covering sources, dataset discovery, DECP search, and supplier monitoring setup.

    Completeness3/5

    The surface covers core DECP querying, source exploration, dataset discovery, and supplier-watch planning, but it lacks direct content-level searching for BOAMP, SIRENE, or concessions, and only fetches short excerpts rather than full records. Users can work around this somewhat through build_supplier_watch and search_datasets, but some operations remain indirect.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      Not graded
      maintenance
      Exposes French and EU public procurement data (BOAMP + TED) as MCP tools for AI agents, enabling search for tenders, awards, and winner intelligence via typed filters.
      4
      163 PyPI
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Provides AI agents with access to French public registries including company records, property sale prices, housing energy performance, and short-term rental rates via MCP tools.
      -
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to search Moroccan public tenders, inspect consultation dossiers, check regulatory compliance under Décret n° 2-22-431, analyze competitor awards and market barometers, and audit submission checklists.
      MIT