mcpField
F
Fr Judilibre

Fr Judilibre

Remote
pipeworx.ioLegal & Government0Updated 3d ago

About Fr Judilibre

Fr Judilibre is an MCP server for legal researchers and practitioners exploring published French judicial decisions. It supports filtered search, full pseudonymised decision text, and lookup of the French labels behind source codes, with cited-text links parsed where possible.

Key features

  1. 01Search published decisions with filters
  2. 02Retrieve full text and structural zones
  3. 03Resolve taxonomy codes to French labels
  4. 04Parse cited texts into article references

Use cases

  1. 01Find decisions by query and court criteria
  2. 02Read a decision and its cited texts
  3. 03Interpret source codes in search results

MCP support

Exposes search, decision retrieval and taxonomy tools to MCP clients, as a local stdio server or through the Pipeworx gateway. Access to its tools requires a PISTE application key.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://gateway.pipeworx.io/fr-judilibre/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

@pipeworx/fr-judilibre

French judicial decisions (Cour de cassation and the judicial courts it indexes) — search, read a decision's full pseudonymised text and structural zones, and resolve the short codes the source uses (jurisdiction, chamber, formation, type, solution, publication) into their French labels. Each decision's cited texts (visa) come with a best-effort {code, article} parse for chaining into @pipeworx/france-legi's fr_code_article.

Part of Pipeworx — an MCP gateway connecting AI agents to 1866+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Every tool needs a PISTE application key, held as a gateway platform key (never in pack source). Without one the tools say so explicitly instead of returning an empty result.

Tools

  • fr_judilibre_search(query?, jurisdiction?, chamber?, formation?, type?, theme?, publication?, solution?, date_start?, date_end?, page?, page_size?) — full-text + filtered search over the published open-data corpus. Returns decision ids, summaries, highlighted snippets.
  • fr_judilibre_decision(id, resolve_references?) — one decision's full text, structural zones (character offsets into text), visa (cited texts) with cited_articles best-effort linked to france-legi, related decisions (contested/forward/rapprochements), attached files.
  • fr_judilibre_taxonomy(id?, key?, value?, context_value?) — resolves the short codes the other two tools use into their French labels (or the reverse), or lists the taxonomy categories when id is omitted.

Auth

Platform key (PLATFORM_JUDILIBRE_KEY, read from _apiKey) — not wired yet. PISTE registration gates access beyond simple keyless signup: an application must additionally tick the Judilibre API and accept its own CGU (separate from the PISTE account CGU), once for sandbox and once for production — a second click-through, not a second approval step. No identity verification, phone, SIRET or human review was found anywhere in the flow (account creation, application creation, or API subscription) as of 2026-10-16.

Data sources

  • Cour de cassation, Judilibre open data (judicial decisions). Served through the state's PISTE API gateway — never named in a tool description or source field (CLAUDE.md 9b); the route lives only in src/index.ts.
  • Reuse: Licence Ouverte / Open Licence 2.0 (permits commercial reuse with attribution — the same licence already governing @pipeworx/france-legi), plus the PISTE portal's own terms of use and a Judilibre-specific terms of reuse document published by the Cour de cassation. The exact text of that last document could not be retrieved verbatim during this build (the courdecassation.fr page returned no extractable content); its existence and the licence stack above it are confirmed from the API's own OpenAPI spec and data.gouv.fr's dataset page.

Implementation notes

  • Auth header is KeyId: <key> (an OpenAPI apiKey-in-header scheme), not OAuth bearer — simpler than most PISTE APIs, which also offer an OAuth client id/secret.
  • visa entries are { id, url, title } with no structured code/article field. cited_articles recognises two citation phrasings actually seen in Cour de cassation prose ("article 1382 du code civil" and "code civil, art. 1382") and marks anything else parsed: false rather than guessing. The Légifrance article id embedded in url (LEGIARTI…/LEGITEXT…) is always passed through regardless of whether the title parsed.
  • The published open-data corpus mirrors what the Cour de cassation selects for publication, not every decision ever rendered — fr_judilibre_search says so in its completeness_note.
  • Source: Cour de cassation, Judilibre open data (judicial decisions), read per request; data_as_of is the date of the call.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "fr-judilibre": {
      "url": "https://gateway.pipeworx.io/fr-judilibre/mcp"
    }
  }
}
What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/fr-judilibre/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1866+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/fr_judilibre_search \
  -H 'Content-Type: application/json' \
  -d '{"query":"responsabilité du fait des choses","jurisdiction":["cc"],"page_size":3}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/fr_judilibre_search. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

{
  "mcpServers": {
    "fr-judilibre": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-fr-judilibre"]
    }
  }
}

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-fr-judilibre

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Fr Judilibre data" })

The gateway picks the right tool and fills the arguments automatically.

More

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.