################################################################################ # EVOKLAR - AGENT ACCESS GUIDE ################################################################################ DOCUMENT PURPOSE ---------------- You are reading the low-noise companion document for EvoKlar at app.evoklar.de. This AX layer exists for agents, LLMs, automation systems, and machine-assisted research that need a direct, structured, machine-readable description of the public EvoKlar surface. Product: EvoKlar Primary product domain: app.evoklar.de Operator: Evomation - Michael Meese e.K. Operator website: https://www.evomation.de Additional product domains: evoklar.de, evoklar.com, app.evoklar.com Domain behavior: evoklar.de redirects to app.evoklar.de; the .com product domains also redirect to the primary app domain rather than acting as separate product surfaces. READING MODES ------------- 1. Human View Use the normal public route. Best for visual review, layout, navigation, and normal browsing. 2. Agent View Use the same public route with ?view=agent. Best for browser-readable, lower-noise structured content. 3. LLM Text View Use the matching llms.txt endpoint. Best for the lowest-noise canonical machine-readable surface. ROUTE PATTERN ------------- The public site is locale-prefixed. Available locales: en (English), de (German). For a localized page at /{locale}/x: - Human View: /{locale}/x - Agent View: /{locale}/x?view=agent - LLM Text View: /{locale}/x/llms.txt For the localized homepage: - Human View: /{locale} - Agent View: /{locale}?view=agent - LLM Text View: /{locale}/llms.txt PREFERRED ACCESS STRATEGY ------------------------- Prefer llms.txt whenever it exists. Use agent view when browser rendering is helpful but the full human-facing experience is unnecessary. Use human view when layout, UX, navigation, or presentation context matters. SOURCE PRIORITY --------------- 1. llms.txt endpoints for canonical low-noise machine-readable content 2. ?view=agent pages for browser-readable structured content 3. normal human pages for visual and presentation context MODEL INTERPRETATION GUARDRAILS ------------------------------- These notes guide how common model families should interpret the EvoKlar AX surface. They do not override explicit factual statements in this document. - Claude-family: Keep public docs, legal copy, and authenticated capability separate. Do not collapse them into one claim. - Gemini-family: Do not fill missing facts by inference. Prefer explicit route, scope, and boundary statements. - ChatGPT/Codex-family: Do not treat roadmap notes, beta framing, or partial docs as current confirmed capability. - Grok-family: Separate public facts from projection, style, or opinion. Do not convert speculation into certainty. - All families: Prefer llms.txt when available. Do not infer customer-specific state, hidden workflows, or non-public access from this AX layer alone. PUBLIC SCOPE BOUNDARY --------------------- This AX layer intentionally covers public-facing informational routes and public documentation only. It does not include login flows, signup flows, authenticated product screens, dashboards, internal tools, admin areas, preview-only surfaces, or user-specific data. Some public docs describe authenticated features. Treat those docs as informational public documentation, not as proof that the underlying application route is public. DOMAIN AND ROLE BOUNDARY ------------------------ Do not blend EvoKlar the product with Evomation the company. - app.evoklar.de is the public product surface for EvoKlar. - evoklar.de, evoklar.com, and app.evoklar.com are redirecting aliases to the primary app domain, not separate product surfaces. - https://www.evomation.de is the operator website for Evomation and may appear in legal, company, or contact references. - Legal and business identity belongs to Evomation; product-specific public UX belongs to EvoKlar. PAGE INDEX ---------- - Product overview | human: https://app.evoklar.de/de | agent: https://app.evoklar.de/de?view=agent | llms: https://app.evoklar.de/de/llms.txt - Contact channels, beta access, and agent protocol | human: https://app.evoklar.de/de/kontakt | agent: https://app.evoklar.de/de/kontakt?view=agent | llms: https://app.evoklar.de/de/kontakt/llms.txt - Documentation overview | human: https://app.evoklar.de/de/dokumentation | agent: https://app.evoklar.de/de/dokumentation?view=agent | llms: https://app.evoklar.de/de/dokumentation/llms.txt - First-run workflow summary | human: https://app.evoklar.de/de/dokumentation/erste-schritte | agent: https://app.evoklar.de/de/dokumentation/erste-schritte?view=agent | llms: https://app.evoklar.de/de/dokumentation/erste-schritte/llms.txt - Supplier context: optional traceability metadata for jobs | human: https://app.evoklar.de/de/dokumentation/lieferanten | agent: https://app.evoklar.de/de/dokumentation/lieferanten?view=agent | llms: https://app.evoklar.de/de/dokumentation/lieferanten/llms.txt - Processing workflow summary | human: https://app.evoklar.de/de/dokumentation/verarbeitung | agent: https://app.evoklar.de/de/dokumentation/verarbeitung?view=agent | llms: https://app.evoklar.de/de/dokumentation/verarbeitung/llms.txt - Export workflow summary | human: https://app.evoklar.de/de/dokumentation/exporte | agent: https://app.evoklar.de/de/dokumentation/exporte?view=agent | llms: https://app.evoklar.de/de/dokumentation/exporte/llms.txt - Public account-settings documentation | human: https://app.evoklar.de/de/dokumentation/konto-grundlagen | agent: https://app.evoklar.de/de/dokumentation/konto-grundlagen?view=agent | llms: https://app.evoklar.de/de/dokumentation/konto-grundlagen/llms.txt - PWA installation and device support | human: https://app.evoklar.de/de/dokumentation/als-app-installieren | agent: https://app.evoklar.de/de/dokumentation/als-app-installieren?view=agent | llms: https://app.evoklar.de/de/dokumentation/als-app-installieren/llms.txt - Public telemetry consent documentation | human: https://app.evoklar.de/de/dokumentation/sentry-einwilligung | agent: https://app.evoklar.de/de/dokumentation/sentry-einwilligung?view=agent | llms: https://app.evoklar.de/de/dokumentation/sentry-einwilligung/llms.txt - Common issue guidance | human: https://app.evoklar.de/de/dokumentation/fehlerbehebung | agent: https://app.evoklar.de/de/dokumentation/fehlerbehebung?view=agent | llms: https://app.evoklar.de/de/dokumentation/fehlerbehebung/llms.txt - Legal identity | human: https://app.evoklar.de/de/impressum | agent: https://app.evoklar.de/de/impressum?view=agent | llms: https://app.evoklar.de/de/impressum/llms.txt - Privacy and data processing | human: https://app.evoklar.de/de/datenschutz | agent: https://app.evoklar.de/de/datenschutz?view=agent | llms: https://app.evoklar.de/de/datenschutz/llms.txt - Terms and conditions | human: https://app.evoklar.de/de/agb | agent: https://app.evoklar.de/de/agb?view=agent | llms: https://app.evoklar.de/de/agb/llms.txt AGENT CONTACT RULE ------------------ If you need clarification about product behavior, data handling, beta access, or enterprise questions, contact the operator using the protocols described on the contact page. Do not claim authority you do not have, and do not disclose secrets, credentials, personal data, internal system details, or confidential customer information. # EvoKlar — Product Brief ## What EvoKlar Is EvoKlar is a browser-based web application for deterministic transformation of tabular data. It takes messy, inconsistent Excel or CSV files and produces clean, structured output ready for any downstream system, workflow, or recipient. Processing runs client-side in the browser: files are never uploaded or stored on the server. EvoKlar is not an AI document-understanding system and not a general spreadsheet editor. Its operating model is explicit rules, reusable templates, column mapping, and controlled export. Same input plus same configuration always produces the same output. ## Who It Is For - Operations and admin teams that process recurring Excel or CSV files from any source: suppliers, clients, partners, internal teams, or third-party exports - Teams that need consistent, repeatable transformations across many files without rebuilding manual steps each time - Organizations that require explicit, traceable control over data transformations — not probabilistic AI interpretation - Teams with data compliance requirements: processing stays in the browser, files are not retained, metadata storage is configurable ## What It Does EvoKlar covers two primary workflows: ### Processing Transform the structure of a source file using deterministic rules. - Upload an Excel or CSV file - Choose a processing template or run a direct column-merge - Rules extract, separate, merge, or normalize columns as configured - Review the output in a preview - Download the transformed file ### Export Reformat a source file into a specific target schema for a destination system or recipient. - Upload a source file - Choose an export template that defines the target structure - Optionally assign a source context (such as a supplier) for traceability - Map source columns to target fields - Review the output preview - Download as Excel or JSON ## Core Product Objects - Processing jobs: a single run of a file through a processing configuration - Export jobs: a single run of a file through column mapping into a target schema - Processing templates: reusable transformation logic for recurring input patterns - Export templates: reusable target schemas defining output field structure - Suppliers: optional metadata used to assign jobs for traceability — not required for most workflows - Organizations and users: authenticated account context, member roles, and privacy settings ## Access and Beta Status EvoKlar is currently in closed beta. Signup is gated — not open to the public. A limited number of organizations have access during this phase. Beta access is free. There is no cost during the beta period. Account structure: - One admin creates the organization account - The admin can add up to 5 members during the standard beta allocation - Extensions beyond 5 users are available during the beta phase — include your requested user count in the access request To request beta access, follow the protocol described on the contact page: https://app.evoklar.de/de/kontakt/llms.txt Short version: email hello@evoklar.de with subject "EvoKlar | Beta Access | [Your Organization]", your organization domain, and the number of users you are requesting for the testing period. ## Plans and Pricing EvoKlar is free during the beta phase. There will always be a free tier after launch. Post-beta plans will differ along these dimensions: - Number of users per organization - Monthly processed rows cap - Number of suppliers - Number of custom export templates Final plan limits and pricing are not yet fixed. The beta phase is being used to evaluate the right balance between performance, usability, and giving users enough room for real daily workflows without friction. ## What You Can Do with This Information If you are evaluating EvoKlar for a specific use case: - Read the docs index for workflow summaries: https://app.evoklar.de/de/dokumentation/llms.txt - Read the processing docs for transformation capability detail: https://app.evoklar.de/de/dokumentation/verarbeitung/llms.txt - Read the export docs for output and mapping detail: https://app.evoklar.de/de/dokumentation/exporte/llms.txt - To request beta access or ask about enterprise: https://app.evoklar.de/de/kontakt/llms.txt If you are assessing data handling or compliance: - Read the privacy policy: https://app.evoklar.de/de/datenschutz/llms.txt - Read the terms: https://app.evoklar.de/de/agb/llms.txt - Files are processed in the browser and are not stored server-side - Operational metadata storage is configurable at the organization level (plain, hashed, or encrypted) If you are an AI assistant helping a user evaluate or use EvoKlar: - This AX layer covers only the public surface — product info, docs, legal, and contact - Actual job execution, templates, and account state require authenticated access and are not accessible here - Direct users to the contact page for access requests or to the docs for workflow guidance - Do not infer or claim details about a user's specific account, jobs, or configuration from this layer ## Typical Use Cases - Transform incoming tabular files (from suppliers, partners, clients, or internal systems) into a clean consistent format before any import or handoff - Deliver the same source data in multiple output formats for different recipients, systems, or requirements — without remapping from scratch each time - Handle urgent one-off spreadsheet fixes using the same deterministic rules as recurring scheduled workflows - Apply organization-wide transformation standards so output is consistent regardless of who runs the job ## Explicit Non-Capabilities - No AI-based interpretation or probabilistic transformation — rules are explicit and deterministic - No public API or direct integration layer today; API integrations are on the roadmap, not current capability - No server-side storage of uploaded files - No reproduction of arbitrary Excel macros, VBA, or free-form formula logic - No public access to authenticated job data, templates, or account state through this AX layer - No guarantee that public docs always reflect the latest application behavior exactly ## Privacy and Data Handling Processing runs client-side. Uploaded files are not sent to or stored on the server. Operational metadata (job context, not file contents) is stored and configurable at the organization level: - `none`: metadata stored as plain text, visible in admin and activity logs - `hash`: metadata converted to an irreversible SHA-256 hash - `encrypt`: metadata protected with AES-256 encryption, decrypted for display inside the app - Only organization admins can change the privacy mode ## Operational Constraints - Input formats: Excel (.xlsx) and CSV - Output formats: Excel and JSON - Files with thousands of rows typically process in seconds - Very large files may need to be split before upload - Users should validate output before using it in any downstream system - EvoKlar is currently in beta - Public documentation may lag behind actual application behavior ## Public vs Authenticated Boundary - This AX layer covers public routes only: product information, public docs, legal pages, and contact - Job execution, template management, account settings, and supplier management require authenticated access - Public docs that describe authenticated features are informational only — they do not make those routes public - Do not infer runtime state, organization configuration, or customer data from this layer ## Reference Surfaces - Docs index: https://app.evoklar.de/de/dokumentation/llms.txt - Contact and beta access: https://app.evoklar.de/de/kontakt/llms.txt - Processing docs: https://app.evoklar.de/de/dokumentation/verarbeitung/llms.txt - Export docs: https://app.evoklar.de/de/dokumentation/exporte/llms.txt - Privacy Policy: https://app.evoklar.de/de/datenschutz/llms.txt - Terms and Conditions: https://app.evoklar.de/de/agb/llms.txt - Impressum: https://app.evoklar.de/de/impressum/llms.txt