Dato Claro · CSV and Excel

Dato Claro · CSV and Excel

v2
by
m
moulharck
Dato Claro · CSV and Excel
Loading...
✦ dato claro
Ready to explore Reset

DATA STUDIO

Turn data into a visual story.

Paste a table, choose a view and find what matters in seconds.

— rows in focus

INTRODUCTION · DECOUPLED DATA ENGINE v1.0

Decoupled data engine

© 2026 Your Name - Guadalupe, N.L.

declarative pipeline

What is it?

Analysis application with a declarative pipeline. Intent is converted into a plan (queryPlan) before processing or rendering.

duckdb-sqlresident
1,570memory
6tiles
256rows / window

↳Does not load everything into RAM.

WHAT IT CAN DO

  • ✓Generate 7 ready-to-share formats: SVG, CSV, Arrow, Parquet, Excel, PDF/A-3, and DuckDB.
  • ✓Embed legal metadata (author, copyright, and license) in each file.
  • ✓Create nested meta-libraries: lib:// → metadata → program/page.
  • ✓Work offline with MHTML/WARC bundles embedded within the PDF.
  • ✓Send any artifact to print from the browser or server (CUPS/Windows).
  • ✓Keep rows private in the browser; only upload aggregated metrics if you enable AI.
  • ✓Weigh less than 10 MB in resident memory despite handling artifacts of 19 MB or more.

WHAT IT CANNOT DO

  • ✕Cannot execute code from a remote library without validation: prevents RCE.
  • ✕Cannot send your original rows to AI or third parties: only aggregates.
  • ✕Cannot print without an installed driver or user confirmation in the browser.
  • ✕Cannot guarantee 100% identical printing in SVG with complex filters: use PDF for that.
  • ✕Cannot function as a real-time collaborative editor: it is a deferred pipeline, not a socket.
  • ✕Cannot be used commercially with third-party ads without a commercial license (PROPRIETARY).
  • ✕Cannot bypass trademark (IMPI) or author (INDAUTOR) registration: you need a process to sue.
  • ✕Cannot convert Arrow larger than 1 GB to Excel: the Excel limit is approximately 1 million rows.

01

Bring your data

Use sample
Format Sheet CSV, TSV, JSON, Excel, and Line-delimited JSON
↥ Drop a file here or choose a file from your device Choose file
Paste your data
Automatically detects columns, dates, and measures Visualize→
Processing file0%
0%

▣Local processing by default: your files are read in this browser. No macros or code are executed; optional AI only receives aggregated metrics.

02

Read code without executing it

Inspect code in an isolated Web Worker and compare it against a uniform BSS representation.

read-only
⌘ Drop a code file Not imported, compiled, or executed Choose code
Paste a snippet or macro
Up to 5 MB · local static analysis
StopAnalyze code→
Analyzing without executing0%
0%

STATIC REPORT

—

not executed
Detected symbols

—

Detected dependencies

—

Detected macros

—

Structure

—

Local multi-agent summaryDeterministic agents summarize, analyze, optimize, and verify the structure without network or execution.
free of charge
Uniform BSS representationA compact sequence to compare structure and behavior across languages.
language agnostic
Structural sequencewithout names or source code
ASE problem-first pipelineModel the challenge as an abstract graph, query signals, and defer interpretation.
local · scoped
Semantic querieson BSS
    Optimization planverifiable
      Signals to reviewNot a security diagnosis

        03

        Local AI reasoning

        Connect Ollama on your machine to coordinate summary, analysis, optimization, and verification without sending code to an external service.

        Ollama · optional
        Local server Installed model Check connection Run local pipeline→ Stop

        ◉The browser only calls your Ollama server. This feature is optional and requires Ollama to be installed and to allow CORS.

        Processing local roles0%
        0%

        Connect Ollama to activate the pipeline.

        04

        Ask for a view

        Describe the chart in simple words. Interpretation happens locally.

        What do you want to compare?
        Apply→
        Trend by date Compare regions Revenue and units

        05

        Decoupled data engine

        Intent is converted into a declarative plan with windows and memory budgets before processing or rendering.

        —

          06

          Your overview

          Chart title ↓
          07
          ARTIFACT LAYER

          Deliver your results

          Export the same analysis in seven formats: engine, human exchange, and offline delivery.

          v2.4 · local History
          LIVE
          pipeline.memory1.570·pipeline.tiles 6·duckdb-sqlresident·query ready
          rows 0mem 0.0 MB
          ◇

          SVG Chart

          vector · editable · current chart

          .svg
          chart view
          Download SVG↓
          ▧

          CSV + frontmatter

          ejemplo.csv · UTF-8

          .csv
          idamountdate
          104212.4302024-11-01
          10438.1202024-11-02
          Export CSV↓
          ▥

          Apache Arrow

          columnar · interoperable · local

          .arrow
          schema3 cols

          int64 id

          float64 monto

          date32 fecha

          Export Arrow↓
          ▦

          Excel + metadata

          2 tabs · metadata

          .xlsx
          SummaryDetail

          Rows processed1.570

          Conversion rate23,4%

          Generate Excel↓
          ✦

          Executive report

          PDF/A-3 · SVG · Arrow · offline

          FEATURED
          Artifactreport + embedded data
          Offlineno internet
          AccessView data · Arrow
          Generate executive report↓
          Includes chart, table, narrative and embedded offline files.
          More formats ▤Compressed Parquet ◈DuckDB Database Improve insights with AI

          If you enable AI, only aggregated metrics are shared; your rows do not leave the browser nor are they used for training.

          Preparing file0%
          0% Stop

          JSON Manifest · traceability

          ready
          {}

          Verified privacy

          Rows remain in the browser. The report can save Arrow, SVG, MHTML, and the manifest inside the PDF/A-3.

          ✓local processing · secure execution

          Recent artifacts history

          {count} artifacts

          —

          Visualization

          Bar Line Area Donut Scatter
          Category Measure
          ⌁Lasso □Rectangle ⊙Point
          ✥Pan and zoom ↺Reset zoom
          Process ⚙Settings

          08

          Data profile

          local analysis

          Load data to see its quality and structure.

          09

          Display settings

          ×

          Customize the canvas without changing your data.

          Accent color Point sizeFrom 3 to 18 px Cache limitFrom 16 to 512 MB
          CancelSave settings
          dato claro Made to look better. Privacy and responsible use
          Local processing

          Your data is read and analyzed in this browser by default; it is not uploaded to a server. Optional AI receives only aggregated metrics.

          Clear limits

          To protect the device, files that are too large and executable formats are rejected.

          Your rights

          You can reset the view and delete saved state from Reset; content of large files is not preserved.

          Usage and license

          Free tool to explore your own data. Current analysis happens locally and does not send data to external models. Verify the licenses of the libraries you integrate.