Skip to main content
Glama

check_citations

Read-onlyIdempotent

Use it to check that a quotation exists, is exact and where it comes from, or to spot an invented one. Finds where a quoted passage comes from (paste the quotation alone, ≤ 3,000 characters, original language). Check each result against the source (link provided). Returns texts and decisions containing it, with a match % (100 = word for word, ignoring punctuation, case, accents) and word differences. Statuses: trouve; trouve_avec_differences; non_trouve (database may be incomplete); trop_court; trop_courant; indetermine (sources timed out: retry). textes_d_origine = statutory sources; repris_dans = decisions quoting it; otherwise decisions, oldest first. Covers codes and statutes (version_du), JO, collective agreements, all French case law, ECHR, CJEU, CNIL, doctrine, professional codes, EU acts in force (note_textes_europeens). Article references in the text are listed, unchecked, with their get_law_article call.

Example: check_citations(text="Les obligations naissent d'actes juridiques, de faits juridiques ou de l'autorité seule de la loi.")

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "text": "<text>"
    },
    "name": "check_citations"
  }
}

Input Schema

NameRequiredDescriptionDefault
textYesthe quotation alone (a sentence or a paragraph), 3,000 characters at most.

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / text / description
      Previous value: -"la citation seule (une phrase ou un paragraphe), 3 000 caractères au plus."New value: +"the quotation alone (a sentence or a paragraph), 3,000 characters at most."
  2. Added
  3. Removed
  4. Changed1 schema field changed
    • changedInput schema / properties / text / description
      Previous value: -"texte libre (projet d'acte, paragraphe, liste de références), 50 000 caractères au plus."New value: +"la citation seule (une phrase ou un paragraphe), 3 000 caractères au plus."
  5. Changed5 schema fields changed
    • addedInput schema / properties / text
      Added value: +{
      +  "description": "texte libre (projet d'acte, paragraphe, liste de références), 50 000 caractères au plus.",
      +  "title": "Text",
      +  "type": "string"
      +}
    • removedInput schema / properties / texte
      Removed value: -{
      -  "description": "Texte libre (projet d'acte, paragraphe, liste de références), 50 000 caractères au plus.",
      -  "title": "Texte",
      -  "type": "string"
      -}
    • changedInput schema / required
      Previous value: -[
      -  "texte"
      -]New value: +[
      +  "text"
      +]
    • changedInput schema / title
      Previous value: -"citations_verifierArguments"New value: +"check_citationsArguments"
    • changedOutput schema / title
      Previous value: -"citations_verifierDictOutput"New value: +"check_citationsDictOutput"
  6. Added

TDQS

A4.6/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, and the description goes well beyond them: it enumerates every status (trouve, trouve_avec_differences, non_trouve with a note the database may be incomplete, trop_court, trop_courant, indetermine with retry guidance), explains the match % semantics, and defines textes_d_origine vs repris_dans vs decisions ordering. The closed-corpus coverage list (codes, JO, case law, ECHR, CJEU, doctrine) is consistent with openWorldHint=false, so there is no contradiction.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The purpose is front-loaded in the first sentence and the closing example makes the call concrete. It is dense — the status and field-name enumeration in the middle is a long run-on — but nearly every clause carries operational information an agent needs to interpret results.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a single-parameter verification tool with an output schema, the description covers input constraints, corpus scope, result field meanings, status interpretation, and failure handling (timeouts, incomplete database). Nothing needed to call it or read its output correctly is left implicit.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds real meaning: only the quotation should be pasted (not surrounding commentary), it must be in the original language, and the 3,000-character cap is restated with a worked example. That goes beyond the schema string 'the quotation alone (a sentence or a paragraph)'.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb and resource ('check that a quotation exists', 'finds where a quoted passage comes from') and adds the discriminating purpose of spotting invented quotations. This clearly separates it from the search_* siblings, which retrieve by topic rather than verifying a verbatim quote.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives explicit invocation conditions: use it for existence/exactness/provenance checks, paste the quotation alone, in the original language, up to 3,000 characters. It also routes article references elsewhere ('with their get_law_article call'), but it does not say when to prefer the search_* tools for a non-verbatim lookup, so an explicit exclusion is missing.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources