Clara AI companion for consultants
Clara organizes sources, surfaces gaps and contradictions, and turns materials, conversations, and data into presentations, documents, and analyses for review. Clara works with you inside Codex.
Create or correct high-impact HTML decks and PowerPoint presentations.
Run interviews and turn recordings into reviewed transcripts.
Map retail attributes and compare new products and best sellers with the rest of the assortment.
Analyze Excel, CSV, and Parquet files with checked calculations and charts chosen to fit the question.
Describe the result you want. You can start from the project materials or an existing presentation.
New presentation
Add Word, PDF, and PowerPoint documents, images, spreadsheets, notes, or recordings. State the audience, objective, and duration: Clara proposes the structure and prepares the presentation.
"Create a 20-minute presentation for this audience from the materials in this folder."Existing presentation
Write the corrections or record them aloud. Clara updates a copy and checks that everything else remains unchanged.
"Record my feedback, apply it to a copy, and leave everything else unchanged."Visual references
Add a PowerPoint template, reference presentation, logo, or brand guidelines. Clara uses them for layout and style.
"Use this PowerPoint as the visual reference for the new presentation."Both formats are editable. Choose PowerPoint when the file must be opened and reworked in PowerPoint; choose an HTML deck for interactivity, navigation, and animations.
A .pptx file that can be opened and edited in PowerPoint. Choose it when you must follow an existing template or hand the file to someone else to edit.
"Deliver an editable PowerPoint and check every slide."A self-contained 16:9 HTML deck with speaker notes, fullscreen mode, keyboard or touch navigation, and animations. It does not require PowerPoint.
"Deliver an interactive HTML deck with speaker notes."Write in your own words: you do not need to learn commands. No separate API key is required; model work uses your existing ChatGPT plan.
Vera and Clara follow the same two-category policy. The distinction depends on the processing boundary, not on the profession.
Work inside Codex
Vera and Clara do not automatically anonymise data. They may use local Python to filter or aggregate information when useful.
Data supplied to the model is processed through the user’s existing ChatGPT plan. Ordinary workflows do not send client files, prompts, or model-context content to Mparanza.
Each workflow is mapped when it is added or changed: what normally stays local and what Codex may read. No record is created for each prompt.
Mparanza-hosted service
Remote Interview, Transcription/Voice Capture, and the retail-data service are hosted by Mparanza. The content needed for those functions reaches Mparanza systems and, where required, OpenAI services.
Each hosted service is documented once at service level, including access and retention or deletion arrangements. There is no prompt-by-prompt documentation.
Watch the data-handling videoClara organizes sources, distinguishes facts, inferences, and professional judgment, and surfaces gaps and contradictions. You decide what goes into the client materials.
Clara makes it clear what comes from the documents and what comes from interpretation or professional judgment.
Clara keeps missing information, contradictions, and pending decisions visible.
Reopen the same folder in a new Codex session and continue from the materials, notes, and drafts already there.
Retailer Signals compares new products and best sellers with the rest of the assortment. Brand Fit relates those signals to the brand's presence and catalogue. Retail data and reviewed mappings use a Mparanza-hosted service; product images and generated reports are saved in your project folder.
Use the data already available in the Mparanza-hosted retail service. Clara starts with the records for the selected retailer and category. The installed version does not yet collect new data from retailer pages.
Standardize and verify the attributes. Clara maps the values to a common taxonomy; a second review checks the mappings.
Compare the groups. New products are those the retailer presents as new. Clara compares new products and best sellers with the rest of the assortment using checked calculations.
Create the report. Clara saves an HTML report on your computer and checks its calculations, readability, and appearance.
For one retailer and category, compare new products and best sellers with the rest of the assortment and generate a verified report.
It starts with a verified Retailer Signals analysis and compares the assortment findings with the brand's presence at the retailer and its catalogue. It uses stored data, not a live shelf check. The comparison is presented in a report.
Clara conducts interviews, transcribes recordings, prepares working documents, and analyzes business data.
Remote interview
The participant opens a Mparanza-hosted link without signing up. Mparanza and OpenAI process the audio and answers to conduct and transcribe the adaptive interview. The interview record remains hosted until manual or administrative deletion; link expiry is not deletion.
"Prepare a 15-minute interview in Italian for this participant."Transcription
Record the conversation or upload an existing audio file through Mparanza-hosted Voice Capture. The audio reaches Mparanza and OpenAI for transcription. Hosted working files are deleted before the completed bundle is returned; Codex may then read the local transcript for speaker attribution and review.
"Transcribe this recording and identify the speakers."Advisory work
Using the documents, notes, and conversations in the project folder, Clara prepares a first draft as a Word document or presentation and flags what is missing before review.
Sources, open questions, and versions are stored in the project folder. Content Codex interprets may enter the model context through your ChatGPT plan.
"Organize the sources, identify what is missing, and prepare a Word document for review."Business data
Start with an Excel, CSV, or Parquet file and describe the business question. Clara chooses the analysis, runs checked calculations, creates the most useful chart, and explains the result.
"Analyze this spreadsheet, choose the most useful chart, and explain the result."Short guides to Clara's main capabilities, narrated in the language of this page.
You can find Clara in the OpenAI Marketplace. After installation, open the project folder in Codex and describe the result you want to achieve.
"Use Clara in this folder."
Not signed in? ChatGPT asks you to sign in, then opens Clara's listing.
