insourcia

get_financials

Read-onlyIdempotent

Historique financier detaille d'une entreprise sur plusieurs exercices.

Defaut plan-aware :

Plan free : mode compact (~40 champs / exercice). Compte de resultat complet (CA -> resultat net en passant par EBITDA, REX, financier, exceptionnel, IS), bilan abrege PCG (actif immobilise net, stocks, creances clients, disponibilites, total general actif, total actif ; capital social, reserves, report a nouveau, capitaux propres, provisions, dettes financieres, dettes fournisseurs, dettes fiscales/sociales, total dettes, total passif), ratios (tresorerie, dette nette, BFR, marges, ratio endettement, CAF, delais paiement), dividendes verses, effectif moyen.
Plan pro : mode full par defaut (~140 champs / exercice, audit financier exhaustif). Override explicite via detail=compact si on veut la vue resumee.

Mode detail=full (audit financier exhaustif) : retourne TOUS les champs financiers disponibles (~140 par exercice). Sur plan gratuit, renvoie 403 upgrade_required ; sur plan Pro c'est le defaut.

Mode fields (recommande pour 1-5 ratios additionnels au-dessus de compact) : passer fields=["roe","bfr_jours_ca","autonomie_financiere"] ajoute les champs cibles a chaque exercice sans gonfler la reponse. Plus de 130 champs disponibles : ratios (roe, taux_marge_brute, liquidite_generale, capacite_remboursement, etc.), postes detailles (achats_marchandises, salaires_traitements, etc.), immobilisations brutes (terrains_brut, constructions_brut, etc.), reserves (reserve_legale, primes_emission_fusion_apport, etc.), croissance (cagr_ebitda_3ans, cagr_rn_signed_5ans, etc.).

Bloc ifrs : pour les societes cotees, retourne en plus un objet ifrs avec les agregats comptes consolides (chiffre_affaires, ebitda, bpa, dividendes, etc.).

RENDU DETERMINISTE — toujours utiliser _layout retourne dans la reponse :

ORDRE : iterer _layout.sections.<section>.lines dans l'ordre fourni. NE PAS inventer l'ordre PCG.
LABEL : afficher line.label (FR humain : "Chiffre d'affaires net", "Marge brute"...), pas line.key brut.
VALEUR : lire exercices[year][line.key]. Si null/absente : afficher "-" SANS supprimer la ligne (les sous-totaux restent visibles).
INDENTATION : 2 espaces par niveau au-dela de 1. level=1 sans indent, level=2 precede de " " (les "dont ...").
EMPHASE : gras pour kind=subtotal|total. Soulignement superieur pour kind=total.
FORMAT NOMBRES (convention FR) :
- Echelle automatique : si max(|values|) > 1 milliard EUR -> afficher en M EUR (3 decimales), sinon en k EUR (1 decimale).
- Separateur milliers : espace insecable.
- Negatifs : parentheses, ex (1 234).
- Pourcentages (label se terminant par "(%)") : 1 decimale.
EXCEL/CSV : 1 sheet par section. Col A = label avec indentation, Col B+ = annees croissantes (ancien gauche -> recent droite). Bold subtotal/total. Format nombre Excel : "#,##0;(#,##0);-".
VERIFICATIONS automatiques (mentionner si KO) :
- bilan_actif.total_actif ~= bilan_passif.total_passif (tolerance 1%).
- Si _layout.not_applicable_pcg: true (bilan B banque ou A assurance) : afficher uniquement les sections presentes + note "Plan comptable sectoriel non disponible".
- Si _layout.missing_pcg_lines.length > 0 : mentionner en bas "Note : N lignes PCG non disponibles dans notre source de donnees : ".

Toujours afficher en sections separees (compte de resultat + bilan actif + bilan passif + ratios). Les postes en lignes, annees en colonnes de gauche (ancien) a droite (recent). Jamais l'inverse.

Input Schema

TableJSON Schema

Name	Required	Description
`siren`	Yes	SIREN a 9 chiffres
`years`	No	Nombre d'exercices (defaut 3, max 10)
`detail`	No	compact (~40 champs, defaut sur plan free) ou full (~140 champs, audit exhaustif, defaut sur plan pro). Sans valeur, le serveur applique le defaut du plan de l'utilisateur.
`fields`	No	Champs financiers supplementaires a injecter dans chaque exercice. Exemple : ['roe','bfr_jours_ca','autonomie_financiere']. Limite par le plan (3 sur free, illimite sur Pro).
`context`	Yes	Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): "Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization."
`type_bilan`	No	K (consolide), C (complet/social), S (simplifie). Sans filtre : meilleure priorite (K > C > S)

Output Schema

TableJSON Schema

Name	Required	Description	Default
`ifrs`	No
`siren`	Yes
`_layout`	No
`message`	No
`exercices`	No
`_user_plan`	No
`denomination`	No
`upgrade_hint`	No
`fields_skipped`	No
`dernier_exercice`	No

TDQS

A4.6/5.0

Behavior5/5

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

Annotations only declare readOnlyHint/idempotentHint/destructiveHint. The description goes far beyond: the 403 upgrade_required behavior on free plan for full mode, plan-aware defaults, the ifrs block appearing only for listed companies, and the deterministic `_layout` rendering contract (order, labels, indentation, number formatting, verification checks). This is rich behavioral context fully consistent with the read-only annotations.

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 description is long but well-structured and front-loaded: mode/plan semantics come first, rendering rules second. The 8-item rendering spec is verbose (Excel number formats, non-breaking spaces) and partially redundant with the output schema's job, but it is organized, numbered, and almost all of it earns its place for an agent that must render results deterministically.

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 tool with 6 parameters, plan-dependent behavior, and a rendering contract, everything an agent needs is present: error behavior (403 upgrade_required), field-count expectations, output structure (_layout sections/lines), formatting conventions, and verification steps. An output schema exists, so return values need no further explanation. No meaningful gaps.

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 baseline is 3. The description adds real value: it distinguishes compact (~40 champs) from full (~140 champs), explains plan-dependent defaults for `detail`, and illustrates the `fields` parameter with a concrete example (["roe","bfr_jours_ca","autonomie_financiere"]) plus the 130+ available field categories. It complements rather than repeats the schema.

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 opening sentence, "Historique financier detaille d'une entreprise sur plusieurs exercices," states a specific verb (get), resource (financial history), and scope (multiple fiscal years). This clearly differentiates it from siblings like get_company, get_credit_risk, and get_directors, so an agent can select it correctly without opening the schema.

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?

The description gives clear, explicit context on mode selection: "Mode `fields` (recommande pour 1-5 ratios additionnels au-dessus de compact)", plan-aware defaults (compact on free, full on pro), and the explicit `detail=compact` override. It doesn't explicitly name sibling-tool alternatives or exclusions, but the mode-level usage guidance is concrete and actionable.

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

All insourcia tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.3/5.0

Disambiguation4/5

Most tools have clearly distinct resource+action purposes, and the descriptions explicitly contrast near-neighbor tools like search_companies vs resolve_companies and get_events vs search_events. The only real risk is the trio of director-oriented tools, especially search_director_companies vs search_companies with dirigeant filters, which requires careful reading to avoid misselection.

Naming Consistency5/5

All 18 tools follow a consistent snake_case verb_noun pattern: search_, get_, list_, create_, watch_, unwatch_, mark_, resolve_. Singular names are used for single-entity actions and plural for list/search operations, making the pattern predictable. There is no camelCase, vague verb, or style mixing.

Tool Count4/5

18 tools is slightly above the ideal 10-15 range, but the count is justified by the broad domain covering search, company intelligence, watchlists, and news. Each tool appears to earn its place, and there are no obvious stubs or redundant duplicates. It feels a bit heavy but not bloated.

Completeness3/5

The company intelligence surface is very complete: search, deep company data, financials, directors, group graphs, events, and credit risk are all covered. Watchlists also have full add/remove/list coverage, but saved searches have a notable lifecycle gap—create and list exist, yet there is no update, delete, or alert-toggle for existing saved searches, creating a management dead end.

Resources

Need Help?