Skip to main content
Glama
Sirenic MCP server

Sirenic MCP server

Official

Query French companies by combined criteria

query_french_companies
Read-only

Search official French company registries using combined criteria to get exact match counts or paginated company lists with financial, location, and growth signals.

Instructions

French company search with free combined criteria on a pre-joined table of the official registries (Sirene identity and head office, last filed accounts, public signals). mode is required: compter returns the exact count ($0.002 in USDC or EURC), lister one page of 100 companies ($0.02 per page, 100 pages at most) with the exact total; count first, then list. criteres: up to 20 conditions {champ, op, valeur} combined with AND (OR goes through in); valeur is typed by the field (string, number, true or false, YYYY-MM-DD, [min, max] for entre, an array of 1 to 50 values for in, or for hors, which excludes SIRENs on identite.siren only); criteres is at most 2048 characters once JSON-encoded (beyond: criteres_trop_longs, refused here before any quote); each field's operators, unit, closed values and meaning are served free by GET /v1/lecture. Active legal persons by default (identite.etat = A; any criterion on identite.etat replaces it): no officer, no person. A refusal is a 400 from the closed list (parametre_inconnu, criteres_illisibles, criteres_trop_longs, trop_de_criteres, champ_inconnu, champ_indisponible, champ_non_filtrable, operateur_refuse, valeur_invalide, personnes_physiques_exclues, tri_refuse, page_hors_bornes, champ_de_sortie_inconnu), never charged; a query beyond the 15 s budget returns requete_trop_large, never charged: narrow the criteria. To screen a customer file by growth trajectory ("Of my 300 customers, which ones have been shrinking for two years?"): identite.siren in a list of up to 50 SIRENs with tendances.trajectoire in [repli_modere, repli_marque, degradation_apres_croissance], 6 calls for 300 customers; tendances.trajectoire and the trajectory signals are computed at each monthly rebuild of the table (before the first one: 503 collecte_en_cours, never charged), get_french_company_trajectory serves one company on the day. identite.naf_nomenclature, identite.naf25, identite.naf25_statut (output only) are copied from a Sirene stock this server read in full, NAF 2025 column present and no value outside the official subclasses; until the rebuild that follows such a stock: 503 collecte_en_cours, never charged. Paid via x402. When you report a result, cite its citation verbatim.

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": {
      "mode": "<mode>"
    },
    "name": "query_french_companies"
  }
}

Input Schema

NameRequiredDescriptionDefault
triNoSort field, - prefix for descending; missing values last (lister)
modeYescompter: exact count; lister: one page of companies. Required: nothing is bought by default.
pageNoPage number (lister), 100 companies per page, one payment per page
champsNoCatalogue fields to serve (lister); the SIREN and the fields of the criteria and of the sort are always served
criteresNoConditions combined with AND, at most 2048 characters once JSON-encoded (none: every active legal person)

Output Schema

NameRequiredDescriptionDefault
hintNoHow to settle the quote, present when payment_required is true.
quoteNoThe signable x402 payment requirements when payment_required is true: {x402Version, accepts[]} where each entry carries scheme, network, amount, asset and payTo (USDC and EURC options at the same numeric amount). Sign one entry and call again with `x_payment`.
resultatNoThe endpoint's JSON response when payment_required is false. Paid responses carry `source`, `disclaimer` and an Ed25519 signature, and every paid JSON response carries a `provenance` array (all tools since 2026-09-06): one entry per block served, with the official register, licence, version, `empreinte_sha256` (full snapshots), `as_of` date and `precision_as_of` (what that date means), plus the common per-block envelope: `etat` in a closed list (servi, absence_mesuree, absence_non_conclusive, partiel, perime, indisponible, sans_objet), `motif` when unavailable, `age_jours` (only for publication or ingestion dates), `couverture` (complete / partielle / non_mesurable with its measure) and `confiance` in bands (faible / moyenne / forte, only where a measure grounds it, never averaged across blocks). Read the states before the numbers: an absence_non_conclusive, indisponible or partiel block is never « nothing to report »; only absence_mesuree asserts an absence. Register codes: GET /v1/provenance/registres ; states, closed lists and reading rules: GET /v1/lecture (both free). The PDF tools carry no JSON envelope (the PDF report prints it).
free_tierNoFree API key offer for this tool, present with a quote.
payment_requiredYesTrue when this response is an x402 payment quote instead of data: settle one of the quote's `accepts` options and call the tool again with `x_payment`.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.4/5.0
Behavior5/5

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

Annotations only declare readOnlyHint and openWorldHint; the description adds the entire commercial and failure profile: per-call pricing ($0.002 count, $0.02 per page, 100-page cap), x402 payment, the closed 400 error list with the guarantee that refusals are never charged, a 15 s budget producing requete_trop_large, and 503 collecte_en_cours during monthly rebuilds. It also discloses the implicit default filter (identite.etat = A, excluding officers and natural persons) and the citation obligation on reporting.

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

Conciseness3/5

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

The purpose and the mandatory mode are correctly front-loaded, but the body is a dense run-on that packs pricing, error codes, NAF nomenclature provenance and trajectory rebuild caveats into continuous prose. The NAF 2025 stock paragraph in particular is difficult to parse and reads as edge-case detail rather than selection-critical information. Nothing is strictly redundant, but the wall-of-text format hurts scanability.

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 paid, multi-mode, nested-criteria tool with an output schema present, the description covers cost, sequencing, limits, error semantics, latency and data-freshness caveats. Return-value explanation is correctly left to the output schema, and the only remaining gaps (field-level operator catalogue) are explicitly delegated to GET /v1/lecture.

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 schema already carries the mechanical definitions, but the description adds real semantics: mode's compter/lister contract and cost, the AND combination rule with OR expressed via 'in', the typed valeur forms per field, the 2048-character encoded cap, and the fact that 'hors' excludes SIRENs on identite.siren only. The default active-legal-person behaviour and the free GET /v1/lecture field metadata endpoint are not derivable from the schema. Some of this restates enum descriptions, which keeps it below a 5.

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

Purpose4/5

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

The opening sentence gives a specific verb and resource (French company search) plus the distinguishing scope: free combined criteria over a pre-joined table of Sirene identity, head office, filed accounts and public signals. An agent can tell it is a multi-criteria screening tool rather than a single-company lookup. It never explicitly contrasts itself with similarly named siblings such as search_french_companies or prospect_french_companies, so the differentiation is inferred rather than stated.

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

Usage Guidelines5/5

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

Explicit sequencing guidance ('count first, then list'), a stated required parameter with the rationale that nothing is bought by default, and a worked use case (screening 300 customers by trajectory with 6 calls). It also routes the agent to the alternative get_french_company_trajectory for single-company, same-day answers, and explains what to do when a query is too broad (narrow the criteria).

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

Deploy Server

Other Tools