Skip to main content
Glama
Sirenic MCP server

Sirenic MCP server

Official

French KYB batch (2 to 100)

get_french_company_kyb_batch
Read-only

Retrieves batch KYB files for 2–100 French companies by SIREN, including commercial-court decisions, enabling compliance checks on supplier or customer lists.

Instructions

Batch KYB: full KYB files for 2 to 100 French companies in one call, commercial-court decisions (Judilibre) included since 2026-09-24. Billed per company at $0.105 (30% off the $0.15 unit price) via x402: the amount is the unit price times the number of SIREN. A SIREN with no diffusible company is returned with trouve=false and billed as one lookup. A SIREN whose identity is incomplete (administrative status or legal category missing) comes back trouve=false with motif identite_incomplete, billed as one lookup. A batch holding such a SIREN (sole proprietorship, restricted diffusion, group without legal personality) is refused whole: 400, never charged, sirens_refuses gives each refused SIREN and its code; replay without them. Each file carries its own per-block provenance (official register, as_of date). Ideal for prospecting and compliance agents processing lists. For a customer or supplier list, pass vue=synthese: the full files of a large batch do not fit in a tool result. One compact line per SIREN (status and cessation date, insolvency proceedings and deregistrations, sanctions screening status and match count, never a name, court-decision count, last filed accounts, completeness, reserves), same checks, same price, batch-level provenance; the text result is compact JSON on a single line. procedure_le is the BODACC publication date of the latest proceedings notice, never the judgment date; radiation_le, the BODACC publication date of the latest deregistration notice, never the date of a judgment or of the deregistration. By default, Claude Code (2.1.293, measured on 2026-10-14) passes the model a tool result of up to 25,000 tokens counted by the API (a result of up to 50,001 characters passes uncounted); beyond that, the model only receives an error message and the path of a file. At the bounded worst case (each field of the lines and of the batch head at its bounded longest), 56 lines are guaranteed to fit, on a strict bound of the exact count: 1,848 tokens for the batch head and 395 per line as counted, each character changed since taken as one more token, at most 1,862 and 409. With real companies, measured on 2026-10-08: 100 SIREN took 16,698 tokens and 80 took 13,754 (2.1 characters per token). A full-size request quotes up to $10.50, above the $1.00 single-payment cap that x402 clients apply BY DEFAULT since @x402/core 2.23 (spendControls): raise spendControls.maxAmountPerPayment, or set spendControls: false, before signing; otherwise your own client rejects the quote without ever calling us. Partial diffusion (rule diffusion_partielle_nom_retenu_et_compte): the file of a unit not proven to be a legal person has identite.denomination null, counted in identite.ecartes.diffusion_partielle. 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": {
      "sirens": "<sirens>"
    },
    "name": "get_french_company_kyb_batch"
  }
}

Input Schema

NameRequiredDescriptionDefault
vueNocomplet (default): the full KYB file of each SIREN; synthese: one compact line per SIREN, same checks and price.
sirensYesList of 2 to 100 nine-digit SIRENs

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

A4.5/5.0
Behavior5/5

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

Far beyond the readOnly/openWorld annotations: it discloses per-company billing at $0.105, that a non-diffusible or identity-incomplete SIREN is still billed as one lookup, that a batch containing such a SIREN is refused whole with HTTP 400 and never charged, and that sirens_refuses enumerates the offenders. It also warns that the default x402 spendControls cap of $1.00 rejects a $10.50 quote client-side before any call — a failure mode the agent could not otherwise anticipate.

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

Conciseness2/5

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

Purpose and price are correctly front-loaded, but the text is one undifferentiated paragraph that drifts into measurement minutiae ("1,848 tokens for the batch head and 395 per line… at most 1,862 and 409", "2.1 characters per token", client version numbers and measurement dates). Those figures do not change the agent's decision beyond "a full-size batch may exceed the result limit"; the field-definition clause for procedure_le/radiation_le is useful but buried mid-stream.

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 high-stakes, billed batch operation it covers everything an agent needs: cardinality bounds, cost arithmetic, refusal and charging semantics, output-size behavior for both vue modes, and the citation requirement. An output schema exists, so it correctly avoids describing return structure beyond the size/shape caveats that are not expressible in the schema.

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

Parameters5/5

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

Schema coverage is already 100%, so the baseline is 3, but the description adds real semantics: it explains what vue=synthese actually returns field-by-field (status, cessation, insolvency, sanctions match count, never a name, court-decision count, accounts, completeness, reserves), that both variants carry the same checks and price, and that sirens_refuses maps each refused SIREN to its code.

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?

"Batch KYB: full KYB files for 2 to 100 French companies in one call" states a specific verb, resource, and cardinality bound, which cleanly separates it from the singular get_french_company_kyb_file sibling. The description also names the included content (Judilibre court decisions) so the agent knows the payload scope.

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 conditional routing — "Ideal for prospecting and compliance agents processing lists" and "For a customer or supplier list, pass vue=synthese" because full files don't fit in a result — plus recovery guidance after a 400 ("replay without them"). It stops short of naming the single-company sibling as the alternative for one-off lookups, so the alternative set is implied rather than stated.

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