Skip to main content
Glama
Sirenic MCP server

Sirenic MCP server

Official

French default risk score, batch (2 to 100)

get_french_company_default_risk_batch
Read-only

Score and rank 2 to 100 French companies by default risk in one batch call, returning risk classes, confidence, and model-version details for customer or supplier exposure monitoring.

Instructions

Use when asked to rank customers or suppliers by default risk, e.g. "Rank my customers by risk and monitor the 30 most exposed." Batch default-risk score for 2 to 100 French companies in one call: each line carries the SAME score, class, confidence and model version as get_french_company_default_risk (one scorecard), in a closed-list basket: expose, a_surveiller, sain, non_evaluable (with its cause), cessee (a ceased company always goes there, whatever its class). Lines are ordered by basket, then by exact axis set (jeu_axes, index into jeux_axes: two scores computed on different axes are never ranked against each other), then by score; motifs_principaux names the at most three scorecard axes that weigh most; a line whose reference year is not exploitable carries qualite_exercice_reference (fiabilite, anomalies): read it before the score. The head counts the non_evaluable lines per cause (non_evaluable_par_cause). An unknown SIREN is a billed line (trouve=false, cause siren_inconnu), and so is one whose identity is incomplete (cause identite_incomplete, with champs_absents and fiche); no line scored at all: 404, not charged, lignes naming each cause (error identite_incomplete when every SIREN is incomplete); BODACC down: the whole batch is cancelled (503), not charged. No officer, no free text; the name only for a legal entity. Statistical indicator about the company, not a decision or an assessment about a person. chainage.surveillance lists the WHOLE expose basket in served order (axis set, then score; 100 at most, never a ceased company) with the 30-day quote of create_surveillance_watch for that list: $0.05 per target x targets. To monitor the first N only, pass its first N SIRENs to create_surveillance_watch (quoted N x $0.05): a separate watch, paid on its own, checked daily, never real time. 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. Paid via x402 ($0.07 per SIREN in USDC or EURC, 30% off the unit score). Closed lists and reading rules: GET /v1/lecture (free). A full-size request quotes up to $7.00, 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. 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_default_risk_batch"
  }
}

Input Schema

NameRequiredDescriptionDefault
sirensYesList of 2 to 100 nine-digit SIRENs (deduplicated; billed per SIREN)

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

A4.7/5.0
Behavior5/5

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

Beyond readOnlyHint/openWorldHint, it discloses pricing ($0.07 per SIREN, 30% off the unit score), which outcomes are billed versus not charged (unknown SIREN billed, 404 and 503 and 400 not charged), and the x402 $1.00 spendControls cap that would otherwise reject a full-size quote. That is behavior an agent cannot get from the annotations or schema.

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 trigger and core purpose are front-loaded, but the body is a dense run-on mixing return-field names (jeu_axes, motifs_principaux, qualite_exercice_reference), billing, and payment-client configuration. Much of the output-field detail duplicates the output schema, and the wall of jargon hurts scannability.

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, batch, open-world scoring tool the description covers the operation, the closed-list result model, error and refund behavior, and the payment preconditions. Nothing an agent needs in order to call it correctly and interpret the cost is missing.

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% and the single sirens parameter is already documented, so the schema carries most of the load. The description still adds real meaning: refusal conditions tied to specific SIREN types, deduplication/billing per SIREN, and the consequence of an unknown or incomplete SIREN line.

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 opens with a concrete verb and resource: a batch default-risk score for 2 to 100 French companies, and explicitly states each line carries the same scorecard as get_french_company_default_risk. An agent can separate this batch tool from the single-company sibling without inspecting either schema.

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?

It gives an explicit trigger ("Use when asked to rank customers or suppliers by default risk"), names the single-company alternative it mirrors, and routes monitoring to create_surveillance_watch with instructions for monitoring only the first N. It also states failure paths (400 refused batch, replay without refused SIRENs) so the agent knows what to do next.

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