ChatGPT Plugin

Firmenatlas

Austrian company register data

Brand: Firmenatlas

Category: Pending
Primary Subcategory: Pending

Integration details

Description

Research Austrian company register data directly in ChatGPT with Firmenatlas. Search for companies and people, examine a company’s financial metrics and explore its owners and shareholdings. Compare metrics such as annual net income, total assets and equity across fiscal years, or compare companies side by side. Access published annual financial statements as PDFs. Find businesses by location, legal form, trade licenses and financial criteria. At your request, save selected companies to lists in your connected Firmenatlas account. Data coverage is limited to Austria. An existing Firmenatlas Pro account is required.

Integration type: Plugin
Verification status: Not applicable
Platform: ChatGPT
Category: Pending
Primary Subcategory: Pending
Secondary Subcategories: None listed
Brand: Firmenatlas
Access: Account required
First tracked: 2026-11-04
Tool count: 22
Market: US

Agent DiscoverabilityChatGPT Plugin Discovery Score

ChatGPT organic discovery is live

Your score is coming

ChatGPT now suggests Plugins on its own when they match a user's request.Your Plugin Discovery Score measures how often yours appears, and it will show here as soon as it’s ready.

ChatGPT recommends Airtable when a user asks it to update a CRM. — ChatGPT surfaces Airtable because it matches the user's request.

Primary Subcategory

Competitive lineup

No Primary Subcategory assigned yetOnce Firmenatlas has a Primary Subcategory, the Plugins and Connectors it competes with will appear here with their scores.

Tools

22 tools agents can invoke

Add a company to an existing Firmenatlas list by list id and Firmenbuchnummer.
Create a new Firmenatlas list in the current tenant account.
Generate a short-lived download URL for a Firmenbuch document by its document key.
Load financial reports by Firmenbuchnummer: latest year by default, explicit years in one call, cursor/limit for older years. This data tool does not render. Focused company cards need get_company_summary resultId plus this financialResultId. Answer individual facts, financial values and ordinary financial tables directly as ChatGPT text or a native table with fiscal years and sources, without a renderer. Use presentation="financials" only for an explicitly requested interactive financial explorer. For "Is this company profitable?" or profit/loss evaluation, use presentation="profitability". Fetch get_company_financials limit=6 and pass financialResultId. Follow the chart with one or two sentences citing the latest year; an optional year-by-year value table is allowed. By default, questions about how a metric developed, including follow-ups, use render_company_card presentation="trend" without an explicit chart request; do not answer only with a native value table. Use a native chart instead when the Firmenatlas view does not suit the requested analysis. metric=the exact returned financial field ID (e.g. bilanz.bilanzsumme, bilanz.eigenkapital, gewinnUndVerlustrechnung.umsatzerloese or mitarbeiter.gesamt). Fetch get_company_financials limit=6 and pass financialResultId; show only the requested metric. For annual statements use presentation="annual-report" with the requested year and documentResultId from list_company_documents, including follow-up years. Distinguish missing values, zero, annual profit and accumulated balance-sheet profit. Cite the financial period. If summary and report profit figures conflict, disclose the inconsistency and withhold a confident yes/no profitability verdict until reconciled.
Load direct owners and holdings for one company. The output uses the same public fields as the REST ownership endpoint.
Ownership: get_company_ownership_graph depth=2, direction=both, then render_ownership_graph. Edges point from owner to holding. The compact list initially shows direct relationships; deeper branches open on request. The large graph is a desktop-only explicit action. For person owners, get_person_profile supplies other actual holdings, not management roles. Re-root on a returned firmenbuchnummer to explore further.
Fetch normalized register data, latest financial and relationship metrics; no AI-researched company information. includeDirectOwners=true includes up to six current direct owners and shares for general overviews only. Reuse this resultId for focused cards or comparisons. General company questions: get_company_summary with includeDirectOwners=true when available, then render_company_card once with presentation="overview"; it starts collapsed. A mere company mention does not require a card. Answer individual facts, financial values and ordinary financial tables directly as ChatGPT text or a native table with fiscal years and sources, without a renderer. Use presentation="financials" only for an explicitly requested interactive financial explorer.
Return the same German filter/result-field metadata and current facets as the REST company-search endpoints. Reuse exact categorical values from segmente.
Load one Firmenatlas list and its analytics-enriched rows. By default this returns a UI-aligned summary view similar to the Firmenatlas list/research tables. Use view="full" only when you need the complete enriched payload for every list item.
Load actual functions, holdings and connections for a resolved person, including management, procuration, board and representation details. Person questions: resolve search_people unambiguously, load get_person_profile status=current, then render_person_card once, also without an explicit card request. Keep actual functions and holdings separate; a connection is not a management role. Historical records require status=all and must not be called current.
Load the UI-aligned person/company network. Connected people are terminal in one response and can be used as a new root in a follow-up call.
Return the shared monthly REST/MCP quota plus MCP request counts, tool-call counts and active MCP rate limits for the authenticated account.
Discover multiple companies by criteria, not a named-company lookup. get_filter_metadata supplies exact fields, facet values and units. finanzhistorie is limited to 25 companies per page and five years by default. Company discovery/segments/lists: get_filter_metadata, then list_companies_by_filters; answer as ChatGPT text or a native table. Firmenatlas has no interactive research-table widget. State applied filters and fiscal years; distinguish total matches from the displayed subset. Fetch further pages or save lists on request.
List discoverable Firmenbuch documents. For annual statements fetch limit=100 and pass this documentResultId to render_company_card presentation="annual-report", along with get_company_financials financialResultId. Direct downloads use dokumentKey with get_company_document_download_url. Annual statements use the requested year, including follow-up years; omit year only for latest by closing date, not filing date. Never substitute older metrics for a newer PDF without figures. comparePrevious=true only for a requested prior-year comparison; a PDF link alone does not render the card.
List all Firmenatlas lists in the current account and optionally mark whether a company is already contained.
Remove a company from a Firmenatlas list by list item id or by list id plus Firmenbuchnummer.
Render the smallest matching company view from get_company_summary resultId. General overview: presentation="overview", initially collapsed. Profitability: "profitability"; metric trend: "trend" with the exact returned metric field ID. Both charts use get_company_financials limit=6 and financialResultId. Annual statement: "annual-report" with financialResultId (years=[year] or limit=2) and documentResultId (list_company_documents limit=100). Document list: "documents". Answer individual facts, financial values and ordinary financial tables directly as ChatGPT text or a native table with fiscal years and sources, without a renderer. Use presentation="financials" only for an explicitly requested interactive financial explorer. Annual statements use the requested year, including follow-up years; omit year only for latest by closing date, not filing date. Never substitute older metrics for a newer PDF without figures. comparePrevious=true only for a requested prior-year comparison; a PDF link alone does not render the card. After a profitability chart answer in one or two sentences with the year; an optional year-by-year value table below the chart is allowed. For multiple companies use render_company_comparison when its view suits the analysis. Render once from retrieved resultIds, never model-generated figures, without an extra quota charge. Keep focused cards inline; no automatic separate panel. Skip UI for text only requests or a host without UI support. Ask for clarification on ambiguous company or person matches. Choose one primary chart for the requested analysis. When you render a suitable Firmenatlas chart, the chart requirement is already fulfilled: never add a native ChatGPT chart of the same companies, metric and fiscal periods by default. A general request for a chart, a different visual style or a restatement of the trend does not justify a second chart. An optional year-by-year value table below the chart is allowed. If a Firmenatlas view is not suitable (e.g. multiple metrics, indexed growth or aggregation), choose a native chart instead of rendering both. After a Firmenatlas chart, add another chart only for a distinct analysis not covered by that view or when the user explicitly requests an additional or alternative chart. Data availability alone does not mean a chart is displayed: use uiPresentation.displayDescription and its selected metric/periods to identify what the prepared view covers; loadedFiscalYears describe supplied reports, not complete metric coverage. Native charts are also appropriate if the component fails to display or UI is unavailable. Answer directly with findings, brief interpretation, fiscal years and sources. Cite Firmenatlas as a data source. Do not announce or describe the UI, mention renderer/tool details, or say "the interactive Firmenatlas chart shows" unless the user asks about the presentation or needs help using it. Verwende die Ergebnis-ID des vorherigen Datenabrufs. Dieser Aufruf verbraucht kein zusätzliches Kontingent.
For 2 to 5 companies, fetch summaries and get_company_financials (limit=6 or requested years) per company, then render_company_comparison once with resultIds and financialResultIds when its view suits the analysis; otherwise choose a suitable native chart instead. General comparisons: presentation="overview", no metric or years, even if detailed reports were fetched. Specific metric: exact field ID; time series: presentation="trend". Use presentation="financials" only for explicitly requested interactive full financial tables. Default to the latest common fiscal year; periodMode="latest" only when latest-per-company is requested. Reported nettomarge measures profitability rates; currentProfit compares absolute profits. Do not render individual or intermediate company cards. Render once from retrieved resultIds, never model-generated figures, without an extra quota charge. Keep focused cards inline; no automatic separate panel. Skip UI for text only requests or a host without UI support. Ask for clarification on ambiguous company or person matches. Choose one primary chart for the requested analysis. When you render a suitable Firmenatlas chart, the chart requirement is already fulfilled: never add a native ChatGPT chart of the same companies, metric and fiscal periods by default. A general request for a chart, a different visual style or a restatement of the trend does not justify a second chart. An optional year-by-year value table below the chart is allowed. If a Firmenatlas view is not suitable (e.g. multiple metrics, indexed growth or aggregation), choose a native chart instead of rendering both. After a Firmenatlas chart, add another chart only for a distinct analysis not covered by that view or when the user explicitly requests an additional or alternative chart. Data availability alone does not mean a chart is displayed: use uiPresentation.displayDescription and its selected metric/periods to identify what the prepared view covers; loadedFiscalYears describe supplied reports, not complete metric coverage. Native charts are also appropriate if the component fails to display or UI is unavailable. Answer directly with findings, brief interpretation, fiscal years and sources. Cite Firmenatlas as a data source. Do not announce or describe the UI, mention renderer/tool details, or say "the interactive Firmenatlas chart shows" unless the user asks about the presentation or needs help using it.
Render verified get_company_ownership_graph data as a compact relationship list; the large graph is an explicit desktop-only action. Respect text-only requests. Reuse resultId without another data or quota call. Verwende die Ergebnis-ID des vorherigen Datenabrufs. Dieser Aufruf verbraucht kein zusätzliches Kontingent.
Person questions: resolve search_people unambiguously, load get_person_profile status=current, then render_person_card once, also without an explicit card request. Keep actual functions and holdings separate; a connection is not a management role. Historical records require status=all and must not be called current. Verwende die Ergebnis-ID des vorherigen Datenabrufs. Dieser Aufruf verbraucht kein zusätzliches Kontingent.
Resolve a company name or Firmenbuchnummer; this lookup does not render. Ask the user to choose ambiguous matches by name, seat and Firmenbuchnummer. Use list_companies_by_filters for company discovery, not named-company lookup. General overviews continue with get_company_summary and render_company_card; focused questions use the matching financial/document tools.
Resolve a person by name and optional birth year; ask for clarification on ambiguous matches. For functions or holdings continue with get_person_profile status=current and render_person_card once. Simple searches do not count against quota.

FAQs

Firmenatlas ChatGPT Plugin FAQ

How the directory, categories and Discoverability Score work.

Read the methodology

How do I improve Firmenatlas's ChatGPT Plugin discoverability?

The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.

Where is this profile measured?

This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.