Do I need an API key?

No. opendata.paris.fr serves all city datasets anonymously over its Opendatasoft REST API. This MCP defines no credentials and needs nothing configured.

What is the difference between the two plaque datasets?

The main dataset (~2,600 rows) is the city-wide inventory across all centuries and themes, with the material, theme and person fields; the 1939-1945 dataset (~1,200 rows) is a dedicated WWII memorials register with its own fields (identifiant, commemorore, adresse_complete, postal code and coordinates).

Why does siecle accept "18-19"?

Century values are stored as text, and the source uses range labels for plaques spanning two centuries — "18-19" is a stored value you can filter on exactly like "19".

Are the addresses complete enough to geocode?

The main dataset stores the street address and the exact placement (floor, façade) plus WGS84 coordinates; the 1939-1945 dataset adds the postal code (stored under the odd field name "empty") and an xy coordinate object.

Paris Plaques: Commemorative Memorials MCP Connector for Claude

A+

Keyless Paris commemorative-plaque data: the 2,600+ city plaques (century, material, theme, person, arrondissement) plus the 1,200+ dedicated 1939-1945 memorials.

8 tools Official Updated Oct 1, 2026 Official Vinkius Partner

More Details Connect to Claude

Paris's two commemorative-plaque datasets — keyless, straight from the Paris open-data platform (opendata.paris.fr).

What you can do

list_plaques / plaque_lookup / count_plaques — the ~~2,600 city plaques: index (the lookup key), arrondissement (integer 1..20), UPPERCASE address, exact placement, the inscription (retranscription), material (~~20 stored values), title, century (stored text: "12".."20" plus range values like "18-19"), theme, commemorated person and countries
plaques_by_century / plaques_by_arrondissement — fast totals for one century or one arrondissement
plaques_overview — the preset facets: the century split, the top arrondissements and the top materials
list_ww2_plaques / count_ww2_plaques — the dedicated 1939-1945 dataset (~1,200 rows): the inscription, the full address, the address detail, the postal code and the WGS84 coordinates

Who is this for

History enthusiasts, journalists on Paris's memory sites, and anyone following a street to see whose plaques hang on its façades. Century values are stored as text — including range labels like "18-19" — while the arrondissement and plaque-index columns are plain integers; the tools build the right clause type for each.

parisfranceplaquesmemorialsww2heritagekeyless

8 tools expose this connector's capabilities to your AI agent.

list_ww2_plaques

Each row has: the record identifier (identifiant, integer), the inscription (commemore, e.g. "DESNOS Robert" or a longer memorial text), the full address (adresse_complete), the address detail (precision_adresse), the postal code (stored under the field name "empty", e.g. 75011) and the WGS84 coordinates (xy, an object with lon and lat). No filters; page with limit/offset (100 rows max per call). List the 1939-1945 commemorative plaques

plaque_lookup

g. 4975) matches the given value. index_plaque is required; 0 or 1 row comes back with the address, placement, inscription, material, century, theme and the commemorated person. Look up a plaque by its index

count_plaques

20), siecle (stored century text), pays, objet (stored theme) or personalite — the same exact-match filters as list_plaques. A fast total count; use it to compare plaque density between arrondissements or centuries. Count plaques by arrondissement, century, theme or person

count_ww2_plaques

No parameters. Count the 1939-1945 plaques

list_plaques

Each row has: the plaque index (index_plaque, the integer lookup key), the arrondissement number (ardt, 1..20), the street address (adresse, UPPERCASE as stored, e.g. "12 RUE PIERRE ET MARIE CURIE"), the exact placement (emplacement, e.g. "RDC" or "1er étage Façade"), the inscription (retranscription), the material (materiau, ~20 stored values like "cuivre" or "marbre"), the title (titre), the century (siecle, stored as text: "12".."20" plus range values like "18-19"), the theme (objet_1, e.g. "les artistes (théâtre et cinéma)"), the commemorated person (personalite) and the countries involved (pays, comma-separated). Filters: ardt is an integer exact match, siecle/pays/objet/personalite are exact stored values. Page with limit/offset (100 rows max per call). List the commemorative plaques of Paris

plaques_by_arrondissement

ardt is required: the number 1..20 applied as a bare integer filter on the ardt column. Count the plaques in one arrondissement

plaques_by_century

g. "19" for the 19th century or the range value "18-19"). siecle is required. Count the plaques commemorating one century

plaques_overview

No parameters. Overview of the plaque inventory: centuries, arrondissements, materials

Connect to Claude

Subscribe on Vinkius, then add this connector to Claude.ai or Claude Code.

① Claude.ai (web app)

Go to Settings → Connectors → Add custom connector
Paste the MCP endpoint URL below

https://edge.vinkius.com/vk_preview_Bb0xySoiwzvy6E2D4VOQdy7IxWYY2S7522oK4TsM/mcp

② Claude Code (terminal)

claude mcp add --transport http paris-plaques-commemorative-memorials https://edge.vinkius.com/vk_preview_Bb0xySoiwzvy6E2D4VOQdy7IxWYY2S7522oK4TsM/mcp

③ Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "paris-plaques-commemorative-memorials": {
      "url": "https://edge.vinkius.com/vk_preview_Bb0xySoiwzvy6E2D4VOQdy7IxWYY2S7522oK4TsM/mcp"
    }
  }
}

Get full access on Vinkius

The preview token above works for testing. Powered by Vinkius.