API Reference Overview¶
Legiqo GmbH's CheckName Partner API exposes one lean primary decision endpoint, one convenience orchestration endpoint, and supporting endpoints for B2B integrations:
POST /v1/name/assessmentPOST /v1/name/screeningGET /v1/name/screening/catalogPOST /v1/conflicts/searchPOST /v1/trademarks/searchPOST /v1/name/eligibility(deprecated)POST /v2/name/eligibility(deprecated)POST /v3/name/eligibility(deprecated)
Endpoint roles in the business workflow¶
These endpoints do different jobs and should not be treated as interchangeable:
| Endpoint | Main purpose | Best used when |
|---|---|---|
POST /v1/name/assessment |
Return a business-ready risk outcome for a proposed name | You need the lean primary decision API with the simplest response semantics |
POST /v1/name/screening |
Run assessment, company conflict search, trademark screening, and optional domain checks in one call | You want a convenience orchestration endpoint for one-call B2B screening |
GET /v1/name/screening/catalog |
Return stable screening codes, enum values, and English fallback descriptions | You want to build or refresh frontend copy/routing maps for screening responses |
POST /v1/conflicts/search |
Find already registered Swiss company names that are identical or similar to the proposed name | You want to identify registered names that may conflict with the proposed name |
POST /v1/trademarks/search |
Screen for potentially similar registered trademarks in the ingested IGE dataset | You want an additional brand/trademark signal before filing, legal review, or customer handoff |
POST /v1/name/eligibility, POST /v2/name/eligibility, POST /v3/name/eligibility |
Legacy eligibility-style outputs only | You are temporarily supporting an older integration during migration to POST /v1/name/assessment |
This reference documents the integration API, not the end-user workflow at checkname.ch. If you are building a partner product or internal tool, this is the surface you integrate against.
Primary endpoint semantics¶
POST /v1/name/assessmentis the main endpoint for partner integrations.POST /v1/name/screeningis the convenience endpoint for partners who want one bundled screening call.GET /v1/name/screening/catalogis the discovery endpoint for stable screening response codes and enums.POST /v1/name/eligibility,POST /v2/name/eligibility, andPOST /v3/name/eligibilityare deprecated and replaced byPOST /v1/name/assessment.- HTTP
200means the assessment executed successfully. - Business decision outcome is in top-level
overallStatus: CLEAR_TO_FILEMINOR_RISKS_DETECTEDMANUAL_REVIEW_ADVISEDREGISTRY_REJECTION_LIKELY- Persist
traceIdfor support/debug correlation. X-Trace-Idmirrors the same identifier in response headers for header-oriented infrastructure.
What POST /v1/name/assessment actually does¶
At a functional level, the assessment endpoint combines several layers into one response:
- Deterministic naming checks for syntax, legal form, geographic wording, and restricted terms.
- Optional semantic analysis when
checkLevel=fullto interpret descriptiveness, purpose mismatch, translations, and implied regulated activity. - Risk-profile mapping that converts triggered rule codes into
INFO/WARNING/ERRORseverities. - Final aggregation into one top-level outcome in
overallStatus.
This means the assessment endpoint is intended to be your operational "decision engine", not just a rule dump.
What POST /v1/name/screening actually does¶
The screening endpoint is the orchestration layer for partners who prefer one call instead of coordinating multiple endpoint calls in their own backend.
By default it bundles:
- company name assessment
- company conflict search
- trademark search
It can also include optional domain availability checks when explicitly requested.
The response is now decision-first:
overallStatusis the main filing-oriented business outcome.resultStatetells you whether all requested checks completed.drivers[]explains which checks materially influenced the outcome.primaryConcerngives one strongest issue for simple UIs and routing logic.detailscontains compact supporting detail for completed checks, including risk-tier counts and truncated top matches.
This endpoint is best thought of as a bundled decision API, not as a raw passthrough of four underlying endpoint responses. Keep using POST /v1/name/assessment when you want the leanest primary decision API with the smallest contract surface.
What GET /v1/name/screening/catalog actually does¶
The catalog endpoint returns the stable machine-readable values used by POST /v1/name/screening:
- assessment issue codes with default severity, category, and English fallback descriptions
- screening driver codes with source, default severity, and English fallback descriptions
- status, action, check-state, risk, and error-code enum values
Partner UIs should use codes and enum values as routing and copy keys. API message fields are English fallback/debug copy only; localized German customer-facing text should come from your own code-to-copy mapping.
What POST /v1/conflicts/search actually does¶
The conflict-search endpoint is designed to answer a different question: which already registered Swiss company names are identical or similar to the proposed name.
It evaluates similarity using multiple signals, including:
- identity or near-identity
- visual similarity
- phonetic similarity
- subset and token overlap logic
Use it when you need to identify already registered names that may create confusion or filing risk, whether for customer-facing flows or internal review.
What POST /v1/trademarks/search actually does¶
The trademark-search endpoint answers a separate brand-screening question: which registered trademarks in the ingested IGE dataset look textually or phonetically similar to the proposed mark.
It evaluates candidate matches using multiple deterministic signals, including:
- normalized text similarity
- phonetic similarity
- token/subset structure
- optional Nice-class overlap or curated related-class proximity
Use it when you want a trademark-oriented screening step in addition to company-name assessment and company conflict search.
Response model philosophy¶
For partner integrations, it helps to separate three layers of meaning:
- transport success: HTTP status and
error.code - business outcome:
overallStatus - exact machine triggers:
issues[].code,drivers[].code, and catalog values
That separation lets you distinguish "the API call failed", from "the API worked but the name is risky", from "here are the precise reasons why".
Versioning¶
- Paths are versioned under
/v1. - Breaking changes are announced through the changelog and partner communication channel.
Content type¶
- Request:
application/json - Response:
application/json
Authentication¶
All endpoints require X-API-Key.