Skip to main content
Glama
Sirenic MCP server

Sirenic MCP server

Official

French company default risk score

get_french_company_default_risk
Read-only

Score a French company's 12-month default risk (0-100) from filed financial ratios, company age and insolvency rulings, returning components, thresholds and confidence for KYB credit decisions.

Instructions

Company default-risk score (0-100) for a French company at ~12 months: credit risk and insolvency risk scoring from filed financial ratios (structure, profitability, liquidity, net cash, debt service, trend) + company age + a hard BODACC override (active insolvency / liquidation / closure for insufficiency of assets = proven default). Returns score, qualitative band, every component with its threshold, and a confidence level. Decision-support only, NOT a solvency opinion or credit rating. Statistical indicator about the company, not a decision or an assessment about a person. couverture_axes names the axes the scale could evaluate and the silent ones with a closed-list reason (no published ratio, holding lens, single-year trend, uncovered line items, commercial-court decisions not read): a silent axis cannot penalise, so fewer computed axes means possibly understated risk, never a better score. Commercial-court decisions linked to the company are counted in contentieux_informatif (by nature and role, partial coverage: zero never means no litigation; dernier_passage_complet dates the collection's last full pass). One measured cell weighs (since scale version defaillance-v1.10, measured against a placebo): at least one order or judgment to pay as defendant dated within the twelve months before the day of computation adds 12 points for an SME (PME) only, never for a mid-cap, a large company or an unknown category (contentieux_informatif.cellule: count, points, closed-list reason); other decisions do not weigh. echelle.backtest carries the dated measurement of the scale on real French defaults (BODACC opening judgments from 31 December 2025 to 3 September 2026 joined with filed accounts, case-control design): AUC, share of defaults per band with confidence intervals, detection at the 25 and 50 thresholds. These are OBSERVED rates on a sample, never a probability of default for the company you are reading; versions_couvertes names the scale versions the measurement holds for. Fully disclosed legal persons only: a sole proprietorship, a restricted-diffusion unit or a group without legal personality is refused, 400 with the exact code, never charged. Paid via x402 ($0.10 in USDC or EURC). 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": {
      "siren": "<siren>"
    },
    "name": "get_french_company_default_risk"
  }
}

Input Schema

NameRequiredDescriptionDefault
sirenYes9-digit 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

A3.6/5.0
Behavior4/5

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

Annotations only declare readOnlyHint and openWorldHint; the description goes much further, disclosing the hard BODACC override, the 12-point SME-only weighting and its closed-list reasons, partial litigation coverage ('zero never means no litigation'), and that a silent axis can only understate risk. It also discloses the commercial model ('Paid via x402 ($0.10 in USDC or EURC)') and the 400-on-ineligible behavior, which annotations do not cover. It stops short of stating latency or idempotency, but the behavioral picture is unusually complete.

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 first sentence front-loads purpose and scope well, and the output contents are described before the caveats. But the body is one dense block of run-on clauses with backticked field names, and the disclaimer material ('Decision-support only... Statistical indicator about the company, not a decision or an assessment about a person') is partly redundant with itself. Several nested parentheticals could be split or trimmed without losing meaning.

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

Completeness4/5

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

For a single-parameter tool with an output schema and a readOnly annotation, the description over-delivers: it explains what the score aggregates, what `couverture_axes`, `contentieux_informatif`, `echelle.backtest`, and `versions_couvertes` mean semantically, and it flags the citation requirement. The only real gap is that it never positions the tool against the other risk/credit siblings, which matters given a 32-tool catalog.

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

Parameters3/5

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

There is exactly one parameter, and its schema description ('9-digit SIREN') plus a regex pattern already fully documents it at 100% coverage, so the baseline is 3. The description adds that only 'fully disclosed legal persons' are accepted, which is a meaningful constraint on the value, but says nothing further about SIREN format or validation beyond the regex.

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?

States a specific resource and deliverable ('Company default-risk score (0-100) for a French company at ~12 months') plus the exact inputs the score derives from (financial ratios, company age, BODACC override). It is unambiguous what the tool produces, but it never names or contrasts the closest siblings such as get_french_company_credit_limit or get_french_company_industrial_risk, so the agent must infer differentiation.

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

Usage Guidelines3/5

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

It draws real boundaries — 'Decision-support only, NOT a solvency opinion or credit rating' and the refusal cases (sole proprietorship, restricted-diffusion unit, group without legal personality → 400, never charged). However it gives no positive when-to-use framing and no routing guidance toward or away from sibling risk tools, leaving the choice between this and credit_limit/industrial_risk to inference.

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