entyrix-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ENTYRIX_API_KEY | Yes | Bearer token from your Entyrix dashboard (required) | |
| ENTYRIX_BASE_URL | No | Override for staging / self-hosted | https://entyrix.com |
| ENTYRIX_TIMEOUT_MS | No | HTTP timeout per request | 30000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| completions | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_companiesA | Find companies by name (or name fragment) across the pan-European dataset. Returns ranked candidates with registry identifier, country, legal form and status. Use this FIRST when you only know a name; then pass the identifier + country to lookup_company or company_brief. Do not use it to filter by industry or size — that is advanced_search. |
| lookup_companyA | Resolve exactly one company by its national registry identifier in a given country. Handles every identifier shape we carry (SK/CZ/EE IČO, AT Firmenbuch e.g. '187a', SI/LV registry numbers, …). Returns the core record: identity, address, legal form, status, NACE. For an LLM-ready summary prefer company_brief; for the full enrichment layer use get_company_details. |
| company_briefA | Compact, deterministic Markdown brief of one company (≤12 KB): identity, address, industry, latest financials, credit score and recommended limit, cyber score, risk flags (insolvency, tax debt, sanctions), NIS2 status and key people where licensed. The best single call for 'tell me about company X'. Accepts every registry identifier shape we carry (SK/CZ IČO, AT Firmenbuch e.g. '033209m', PL REGON, FR SIREN, …); pass |
| get_company_detailsA | Full record for one company: identity, address, legal form, status flags, NACE, and the whole enrichment layer we hold (web, tech stack, NIS2, EU funds, LEI, HU de minimis state aid granted, HU public procurement from EKR, …). Large — prefer company_brief unless you need a specific field. Natural-person fields are gated and may be withheld; a withheld field is not the same as an empty one. |
| get_financialsA | Filed financial statements of one company, one row per fiscal year (newest first), with revenue, profit, assets, equity, liabilities and year-over-year trend. Amounts are EUR CENTS as integers (divide by 100 for euros). Use for trend and solvency questions; company_brief already carries the latest year. |
| get_company_networkA | Companies connected to this one through shared officers or owners (one hop). Use for group structure and beneficial-ownership tracing. Person names are gated like everywhere else. |
| get_company_relationsA | Direct relations of one company — statutory officers, owners, parents, subsidiaries, M&A events — without traversing the wider graph. |
| advanced_searchA | Structured company search for portfolio building and market sizing — NOT for finding one known company (use search_companies). Filters: country, NACE prefix, status, region, turnover range (EUR), employee range, founding date range, VAT payer, tax debt, sanctions. Results are paged (limit + offset); the response says how many matched. Only the listed arguments exist — anything else is rejected, never silently ignored. |
| kyb_checkA | One-call Know-Your-Business summary for one company: registry status, insolvency, liquidation, sanctions screening, debarment, tax/social/health debts, VAT reliability and the overall risk verdict. Use before onboarding a customer or supplier. Coverage differs by country: |
| screen_companiesA | Run kyb_check on up to 25 companies at once and get one compact row per company (status, insolvency, sanctions, debts, verdict, coverage). INSUFFICIENT_DATA = no findings but not covered by the core registers for that country — never report it as clean. Use for supplier lists or portfolios. Each company counts as one API request against your rate limit. Requires an authenticated account. |
| check_complianceA | Lightweight public compliance snapshot for one company (works without an account): sanctions screening, insolvency, debtor registers, VAT reliability. For the full KYB picture use kyb_check. A null field means WITHHELD or not measured — it does NOT mean clean. |
| find_suppliersA | Public procurement contracts this company is party to (SK CRZ + CZ Registr smluv), newest first, paged. HONEST SCOPE: this is business-to-government procurement, NOT a commercial supply chain — do not present it as one. |
| list_rankingsA | A published ranking by its slug (e.g. largest employers, top taxpayers). Call WITHOUT a key first to list the available slugs — inventing one returns an error. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| kyb_supplier_check | Know-your-business check before onboarding: identity, status, insolvency, sanctions, debts, financial health, with an explicit verdict and what could not be verified. |
| due_diligence | Deeper due diligence before a contract: group structure, officers, financial trend, public contracts and compliance signals. |
| nis2_supply_chain | Assess a supplier for NIS2 supply-chain security: NIS2 scope, cyber posture signals and business-continuity risk. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| agent-playbook |
TDQS
Scored across 13 tools
Most tools target clearly distinct operations, and the descriptions actively route the agent between lookalikes (search_companies vs advanced_search, company_brief vs get_company_details, kyb_check vs check_compliance vs screen_companies). The remaining overlaps are deliberate tiers of the same resource (lookup_company/company_brief/get_company_details, and the three compliance checks), which are well disambiguated by text but still require reading rather than being self-evident from the names.
The dominant convention is verb_noun (list_rankings, search_companies, lookup_company, get_financials, find_suppliers, screen_companies), which is applied consistently across most of the surface. Three names break the pattern into noun/adjective phrases (company_brief, advanced_search, kyb_check), but they remain readable and unambiguous.
Thirteen tools is a well-scoped size for a company-intelligence/KYB domain, with retrieval tiers, search modes, batch screening and procurement all earning a slot. Nothing appears redundant filler; each tool maps to a recognizable workflow step.
The lifecycle is broadly covered: discovery (search/advanced_search/rankings), single-entity retrieval at three depths, financials, ownership/relations, KYB screening (single, batch, unauthenticated) and public procurement. Gaps remain around multi-hop graph traversal (network is explicitly one hop), person/officer-centric lookup, and procurement coverage limited to SK/CZ, but these are minor and workaroundable.