So werden Ihre Daten verarbeitet.

Purpose. Understand each selected client document and propose an evidence-backed studio archive category, practice, date, name, anomaly status, or probable-duplicate relationship

Information. One registered client and engagement identity; local client-relative paths or bound Google Drive root, file and parent IDs; file names, MIME types, hashes or available Drive checksums, sizes, timestamps and versions; selected readable local content or transiently downloaded/exported Drive evidence; proposed category, document type, date, subject, reference, practice, confidence, reasons, anomaly observations, and probable-duplicate relationships

Purpose. Prepare a dry-run filing plan, collect professional decisions, and execute only a separately approved mechanically safe path plan

Information. Versioned studio policy; storage mode; current and proposed client-relative paths; exact-duplicate checksum groups; Drive file, parent and folder IDs where applicable; collisions and blocked reasons; review payload; reviewer alias, actions, notes and edited paths; approved plan; pre-apply local hash or Drive version-state checks; apply and rollback journal state; artifact receipts and execution limitations

Purpose. Map sources, review reconciliations, and explain exceptions

Information. User instructions; the selected Studio Archive customer-folder run context, including stable opaque client, engagement, workflow and run identifiers; exact imported-input or upstream-artifact receipts and run-local execution copies; accounting files or extracted records; mappings; reconciliation results; exception evidence; workpapers, artifact purposes and audiences, lifecycle state, and review decisions

Purpose. Discover and match source-backed opportunities, then prepare a reviewable grant-application dossier after selection

Information. Opaque client references, receipted profile-evidence references and dated revisionable opportunity-profile facets; professionally reviewed priority-source registry metadata for territory, category, official surface and act families; model-led query-scoped source selections with exact provider provenance, one professionally reviewed covered-or-gap claim for every requested territory and category, selected source IDs and rationales; scan windows, immutable source-check snapshots, explicit coverage gaps and optional last-seen publication cursors; opportunity lifecycle and amendment history; client/opportunity matches, gaps, contradictions, economic assumptions and reviewed recommendations; selected-client self-verifiable opportunity handoffs; call, formal amendments, annexes, official FAQs, forms and instructions; applicant identity and corporate, financial, quotation, declaration and project evidence when professionally relevant; bounded intelligence packets; extracted requirements, exact stored excerpts and facts; eligibility, exclusion and cost assessments; document checklist; form and narrative drafts; issues, recorded model suggestions and professional-review decisions

Purpose. Review accounting meaning, mapping candidates, missing disclosures, schedule evidence, narrative drafts, validation issues, and professional decisions for an Italian OIC annual-accounts case

Information. User instructions; entity profile and tax identifier; CSV/XLSX and readable/scanned PDF trial balances and account descriptions; bounded PDF extraction candidates, page/table/cell geometry, OCR text and confidence, proposed headers, professional corrections and exclusions; prior accounts and XBRL, including explicit and typed dimensional contexts and context-fact groups; fixed-asset, receivable, payable, loan, equity, tax, payroll, provision, related-party, guarantee, post-closing-event, and going-concern evidence; source anchors and checksums; mapping candidates and accepted decisions; reviewed schedule-cell taxonomy dispositions and derived facts; the selected OIC professional-review checklist and structured disclosure answers; narrative drafts; validation issues; approval and export metadata

Purpose. Provide workflow guidance, mapping proposals, question prioritization, narrative drafts, prior-year comparisons, and issue explanations inside the existing bilancio process

Information. Task-specific context packets containing only the selected account rows, accepted facts and answers, active questions, prior text suggestions, bounded pending-PDF headers, sample rows, geometry, OCR confidence and extraction issues, bounded statutory-presentation gaps and arithmetic issues, or validation issues needed for the requested semantic task; direct entity identity and the out-of-band case routing identifier are excluded where the task does not require them; model output remains non-authoritative and evidence-linked and cannot accept, correct, exclude, or promote PDF extraction rows

Purpose. Map selected entries and resolve accounting-support checks

Information. User instructions; stable Studio Archive client, engagement and run identifiers; exact same-engagement Journal Sampling normalized-population, diagnostics and sample artifact bindings; exact support input receipts and run-local execution copies; lifecycle and output paths; sampled journal entries and signed amounts; FatturaPA XML or connector exports; supporting PDFs; invoice numbers, dates, currencies, supplier or customer names and tax identifiers; source metadata; match results; evidence excerpts; artifact purposes and audiences; and reviewed party, relationship, currency and direction decisions

Purpose. Classify the incoming file, interpret evidence, and prepare the studio intake

Information. Incoming client documents or extracted text; document names and evidence passages; structured fiscal fields; formal checks; extraction diagnostics; evidence locators; missing items; studio memo, client-email draft, review decisions, and package state

Purpose. Assess whether a professional development warrants communication and prepare source-backed studio-formatted drafts and visual stories for review

Information. Explicitly selected current source documents and official-source snapshots; selected prior studio emails, circulars, website articles, newsletters, FAQs, alerts and social posts; studio name, logo, colors, contact lines, optional format brief and approved or proposed format profile with field-level evidence basis; intended audience and client-segment descriptions; answer contract, source-authority assessments, exact public source notes and registered public URLs, claim-assurance review, editorial-value judgment, atomic claims, temporal qualifications, uncertainty, professional judgment, master brief, channel drafts, visual story, versioned model-pass prompts and SHA-256 digests, operator-attested host session identifiers, editorial-assessor benchmark cases and result, optional creative-direction handoff and selected board-reference metadata, review decisions and packaged artifacts. Real professional names and contact details may enter the selected runtime model context when present in selected material. The workflow does not promise automatic anonymization, provider-authenticated model-call receipts or local-only processing.

Purpose. Identify the concordato preventivo procedure, governing framework, authoritative proposal, plan, attestation, court material, and document perimeter

Information. User instructions; debtor and procedure identifiers; court and filing details; plan, proposal, attestation, professional and court documents; extracted text; document versions, roles, evidence locators, and reviewer judgment bases

Purpose. Reconstruct and review the creditor perimeter, priority, classes, voting treatment, recoveries, timing, disputes, and liquidation comparator

Information. Creditor names and identifiers; claim amounts and status; secured, privileged, unsecured, subordinated, tax, and social-security information; classes; vote and treatment data; proposed and liquidation recoveries; payment timing; evidence locators; professional assessments and follow-up

Purpose. Review sources and uses, liquidity, distributions, milestones, assumptions, consistency, and feasibility evidence

Information. Business-plan and accounting data; assets and disposals; financing and external contributions; procedure costs; cash flows and liquidity; distributions; milestones; assumptions; recalculations; inconsistencies; missing evidence; and reviewer findings

Purpose. Record review questions, issues, evidence requests, reviewer decisions, and professional handoff material

Information. Semantic case-model decisions; judgment bases; open and resolved issues; severity and ownership; requested documents; reviewer notes and actions; generated workpapers, summaries, and review memos

Purpose. Assess answer-contract conformance, claim coverage, source identity, semantic support, reasoning quality, professional-judgment boundaries, and proposed corrections

Information. Generated or supplied answer, including a Deep Research report, memo, or one-page letter; answer contract; claims and surrounding text; citations; normalized captured text from fetched public pages or local source files; source-identity, semantic-support, reasoning, issue-treatment, disposition, and professional-judgment assessments; diagnostics; proposed corrections and review decisions

Purpose. Establish the accounting meaning, perimeter, mappings, relationships, reconciliation posture, and review status needed to prepare financial-analysis and financial due-diligence evidence

Information. User instructions; trial balances, general ledgers, P&L tables, receivables, payables, inventory, revenue and customer files; management adjustments; cash, debt, lease, factoring, fixed-asset and Capex records; working-capital targets; contingent-liability evidence; transaction definitions; EBITDA-to-cash and Enterprise-to-Equity bridge inputs; financial issues and deal-decision references; source identities and hashes; dataset and relationship contracts; reviewed crosswalks and decisions; calculation parameters; metric receipts; reconciliation results; prepared tables; replay evidence; limitations and review status

Purpose. Resolve mappings and review matched, unmatched, or ambiguous transactions

Information. User instructions; bank statements; journals or ledgers; normalized rows; controls; matches; exceptions; supporting evidence and review decisions

Purpose. Classify, identify, or suggest eligible relationships for unresolved bank movements under the user's selected certainty threshold

Information. A bounded packet of unmatched bank rows, hard-compatible journal candidates, normalized dates and amounts, descriptions, beneficiary or counterparty details, references, movement numbers, account and perimeter fields, plus the reviewed matching policy

Purpose. Resolve journal mappings and explain the sampling design and selected sample

Information. User instructions; the stable Studio Archive client, engagement and run identifiers; the exact journal input receipt and run-local execution copy; the customer-folder output root and lifecycle state; journal data or normalized rows; mappings; population statistics; sampling parameters; selected rows; diagnostics, artifact purposes and audiences, and review decisions

Purpose. Prepare identity, engagement, privacy, AI-applicability, AML, document, and monitoring proposals for professional review

Information. Byte-verified file-preparation outputs or registered standalone evidence; identity, representative and beneficial-owner facts; engagement terms; privacy and AML records; source, template, chronology, and evidence passages; calculations; temporal-horizon state; gaps; proposals, review decisions, and package state

Purpose. Assess or create a professional studio website, propose its information architecture, copy and visual system, implement a responsive site, and prepare exact preview and release packages for review

Information. Selected captures of an existing public website; selected source code, screenshots, studio documents, approved communications, logo, photographs, contact details, team and professional biography, verified services and qualifications, supplied legal text, intended audiences, proposed sitemap, page copy, studio-profile fields with observed, user-supplied or Vera-default basis, design tokens, implementation files, browser screenshots, model-led quality assessments, deterministic validation findings, review decisions, package hashes and visible preview or publication receipts. Professional names, portraits, contact details and other real studio information may enter the selected runtime model context when supplied or selected for the website. The workflow does not promise automatic anonymization or local-only model processing.

Purpose. Structure facts, resolve evidence conflicts, and apply supported sources to the case

Information. INPS records and exports or captured page evidence; contracts, F24s, emails and scans; extracted text; page quotes; facts; timelines; official sources; calculation basis; draft findings and review decisions

Purpose. Understand the professional question, define the answer and validation contract, write route-specific generation instructions, and review their semantic conformance

Information. The question and facts typed or attached by the user, including names or other personal facts when material; inferred legal, tax, or compliance framing; answer contract including document type, validation scope, correction policy, and professional-judgment policy; generation route and instructions; model-led prompt-to-question and prompt-to-contract conformance assessments; source-domain sidecars; diagnostics and review decisions

Purpose. Frame the case, apply current official guidance, and draft the professional-review practice plan

Information. Case facts and local evidence; DIRE or SARI screenshots and OCR; dates and chamber; official-source passages and provenance; proposed classifications, recipient positions, filing steps, questions and review decisions

Purpose. Map source tables, write the report narrative, and review the generated report

Information. User instructions; spreadsheets, CSV files or extracted PDF text; table inventories and rows; mappings; calculations; diagnostics; draft narrative; report artifacts and review decisions

Purpose. Interpret and review commercial and FX assumptions, then prepare a forward-looking sales Plan from reviewed Actuals

Information. User instructions; monthly Actual sales rows, including sparse observed grains; product, customer, country, channel or other reviewed dimensions; units, prices, gross sales, discounts, COGS, transaction and reporting currencies, and FX rates; Actual-to-Plan period mappings; reviewed assumptions, scopes, priorities, same-driver overlap behavior, discount and COGS assumption bases, and rationale; source identities and hashes; Plan scenario rows; assumption ledger; scenario summary; reconciliation results; replay evidence; limitations and review status

Purpose. Find and interpret prior studio documents within an explicit archive scope

Information. User questions; the selected connected-file or configured archive scope; relative document paths; search snippets or opened source passages; file, page, paragraph, sheet-row, message-line, or text-line locators; source hashes when available; extraction and coverage limitations; and Vera's source-backed synthesis

Purpose. Bind Vera workflows to one stable registered client and current Studio Archive folder without copying the private identity registry

Information. The stable opaque client ID; current opaque folder scope ID; absolute archive and client-folder paths; archive-relative scope path and display name; and a deterministic content digest. The binding contains no email address, legal name, tax identifier, document content, Gmail content, or authentication material

Purpose. Create or resume any registered client-bound Vera workflow across separate Codex chats

Information. The selected customer-folder manifest and current local or Google Drive folder binding; explicit engagement, workflow and run IDs; source, journal, support, local-folder snapshot, Google Drive snapshot, or upstream-artifact roles; canonical input-receipt and run execution-copy paths; original filenames, bounded client-relative snapshot paths, Drive file and parent IDs, MIME types, versions, capabilities and available checksums, excluded links, shortcuts or special entries, byte counts, SHA-256 receipts and timestamps; transiently opened Drive text and citations; run labels, purposes and lifecycle state; portable run contexts and input and artifact manifests; exact workflow output paths; artifact purposes, audiences and media types; and recovery or retention inventories. Email addresses, legal-name aliases, tax identifiers and OAuth token contents remain in private local state except when the user supplies non-token identity facts in the current chat

Purpose. Find and interpret Gmail correspondence for one explicitly selected studio client

Information. Client name or identifier and full email or PEC addresses supplied or confirmed in the current task; an optional private local archive scope and identity profile only when the selected runtime supports them; bounded discovery and exact-address Gmail queries; connected account profile; candidate and matching message or thread identifiers; returned sender and recipient fields; subjects, timestamps, snippets and labels; selected message or thread bodies; selected attachment metadata or content when the connector supports them; exact-address routing evidence; ambiguity findings; search coverage; and Vera's source-backed synthesis

Purpose. Inspect and interpret visible messages in one verified local WhatsApp Desktop chat for an explicitly selected studio client

Information. The selected client name or identifier and complete international phone supplied or confirmed in the current task; sanitized guarded-search status and a fresh target result index; verified one-to-one chat identity; the exact target ChatMessagesTableView subtree only after phone verification; visible sender or profile name; visible message text or captions, timestamps and on-screen locators; unreadable media, read-state, history and coverage limitations; focus or identity failures; and Vera's source-backed synthesis. Raw pre-verification accessibility snapshots remain inside the local Computer Use JavaScript call so unrelated sidebar previews are not returned to model context

Purpose. Map product attributes and compare retailer cohorts

Information. Retailer and category; structured product, variant, brand, description, price, promotion, ranking, cohort and taxonomy records; public image URLs and locally hydrated product images; package provenance and mapping history

Purpose. Resolve ambiguous taxonomy mappings and author and review a checked report

Information. Mapping tasks, decisions, evidence, local-image hashes, independent mapping reviews, computed tables, report model, claim ledger, caveats, selected examples, semantic review, browser QA, and correctness verdict

Purpose. Compare retailer signals with current retailer presence and the brand-owned catalogue

Information. Completed Retailer Signals report and verdict; retailer, category and brand names; current database snapshot, retailer-presence rows, owned-catalogue rows, accepted mappings, product attributes, candidate products, image URLs and locally hydrated images

Purpose. Interpret gaps and candidates and create and independently check the local report

Information. Scope metrics, deterministic claims, report model, selected products, rationales, caveats, semantic review, browser QA, correctness findings, and final HTML

Purpose. Capture or audit the basis for claims in a presentation

Information. PPTX visible text and shape names; slide titles and claims; source titles, locators and URLs; calculation inputs and methods; reasoning inputs, assumptions, and cross-slide claim references

Purpose. Identify changed, missing, moved, untracked, or unsupported claims

Information. Generation-time claim snapshots, current-deck text, matched support, drift statuses, ungrounded items, and required semantic refresh decisions

Purpose. Understand and organize an advisory engagement

Information. Client and project labels; objectives and audience; company, participant and stakeholder identities; source documents, decks, spreadsheets, notes, transcripts, recordings, dates, facts, financial or operating data, constraints, risks, and open questions

Purpose. Weigh evidence and prepare decision-ready advice

Information. Material summaries, evidence maps, source limitations, consultant and advisor judgement, Codex inferences, contradictions, hypotheses, options, recommendations, implementation conditions, pending inclusion decisions, case issues, and pseudonymous but linkable audit metadata such as exact source digests, byte counts, stable IDs, dates, counts, and cross-receipt hashes

Purpose. Draft and review workpapers, briefs, decks, memos, and decision packs

Information. Narrative, charts, claims, citations, speaker notes, review findings, approved content, and final deliverables

Purpose. Interpret requested changes against the correct deck and evidence

Information. Existing PPTX or HTML deck; slide text, images, notes and structure; spoken feedback, transcript, screen recording, review notes, partner comments, case evidence, style authority, and source materials

Purpose. Plan, apply, and verify only the understood changes

Information. Interpretation packets, requested changes, uncertainty, targets, success criteria, approval state, patches, corrected deck, rendered slides, verification findings, and final audience-facing review

Purpose. Build a source-faithful presentation

Information. Word, PDF, Markdown, spreadsheet, case-workspace, sealed JSON or CSV evidence artifacts, or mixed source material, including names, facts, figures, qualifications, evidence gaps, and advisor judgement

Purpose. Plan, compose, revise, and review the presentation

Information. Storyline, claims, evidence references, evidence bundle and binding records, resolved values, slide titles and text, speaker notes, layout choices, content and evidence ledgers, HTML/CSS, screenshots, and QA findings

Purpose. Select or prepare an interview that addresses the engagement question

Information. Campaign definitions; case and project labels; participant name, role, language, introduction and identifier; purpose, background context, hypotheses, topics, questions, red flags, boundaries, and expiry

Purpose. Retrieve and interpret the completed interview

Information. Status, completion data, transcript events, participant answers, media metadata, generated review, uncertainties, contradictions, missed opportunities, and follow-up questions

Purpose. Profile, interpret, validate, and visualize the user's business data

Information. CSV, XLSX, or Parquet fields and values; dataset identity; metrics, dimensions, periods, identifiers, members, source notes, and compatibility evidence

Purpose. Author reusable dataset semantics and choose a source-backed analysis

Information. Dataset profiles, intake receipts, semantic layers, evidence, reviewed Sales/Discount/COGS mapping states, role bindings, snapshot attachments, analysis policies, chart plans, exact input and output byte hashes, effective recipe records, rendered charts, validation results, and interpretations

Purpose. Transcribe an advisor debrief, meeting, call, or existing recording in the correct project context

Information. Audio or screen-recording content; compact case brief; client and project facts; recording title, date, participants, interviewer, notes, language, and source metadata

Purpose. Attribute speakers, review evidence, and integrate the transcript

Information. Raw and reviewed transcripts, speaker identities or labels, timing and media metadata, uncertainty notes, transcription corrections, review findings, judgement proposals, questions, issues, and evidence-map updates