User Guide

RangerIO is a desktop data-intelligence and sanitization tool for working with files that may contain sensitive information. It imports supported files, scans them for potential PII, provides review and correction tools, answers questions about them, builds a reusable file profile called an Intelligence Dossier when you ask, and prepares sanitized or tokenized context for the AI provider you choose.

In everyday terms, RangerIO helps you open a work file, understand what is in it, check it for sensitive information, ask questions about it, and prepare an output for review or sharing.

What do you need to do?

A good first workflow

RangerIO is not only a chat interface. The data engine works before a model is involved:

1. 1You import a file.
2. 2RangerIO reads its structure and indexes it.
3. 3RangerIO scans it for personal and sensitive values.
4. 4You review the findings.
5. 5You ask questions, build an Intelligence Dossier, create a report, or export a masked copy.
6. 6Your AI provider is used only when the task needs a model, and personal details are replaced before anything is sent.

RangerIO Enterprise

RangerIO Enterprise is a licensing tier, not a separate product. It runs the same data engine, privacy workflows, and export paths described in this guide, and adds two things for organizations: centralized management for multi-seat deployments, and fully supported database and REST API data sources alongside file imports.

• Centralized management for deployment, configuration, and policy across many workstations.
• SQL database sources as a supported, fully backed import path.
• REST API sources as a supported, fully backed import path.
• Full support for the organizations running these connected sources.

What you can do

• Create projects and project groups.
• Import individual files, or a Cowork Folder that RangerIO watches for new and changed files.
• Follow import tasks and recover an interrupted import.
• Preview data and build an Intelligence Dossier for a file.
• Review PII findings in tables and documents.
• Mark missed sensitive values, dismiss false positives, and change the PII type.
• Ask questions about one or more ready files.
• Review grounding, coverage, data-quality caveats, and what was replaced before a cloud request.
• Export data, reports, and masked copies.
• Turn on the local MCP server and connect Claude Desktop.

What needs an AI model

RangerIO does not include an AI model. You connect your own: a cloud provider with your API key, or a local model server you run, such as Ollama, LM Studio, or vLLM. Questions on Explore, the Intelligence Dossier, and AI-assisted reports need a connected model. Import, privacy review, and standard exports do not.

Before relying on an output

A completed import, a “Ready” status, a PII scan, a masked export, or an AI answer is not a certification. Review consequential outputs against the source material and your organization's policy.

Terms you will see

Professional work often starts with files that should not be pasted directly into a general-purpose AI chat: contracts, employee files, accounting workbooks, research exports, operational records, and other client material.

RangerIO separates preparing and understanding the source from choosing an AI engine.

The practical problem

Without a preparation layer, a user has to decide between:

• doing the work manually;
• sharing more source content than intended;
• manually removing identifiers;
• rebuilding the same context in every new AI conversation; or
• using an AI answer without a reusable, inspectable source profile.

RangerIO addresses this by building reusable local intelligence first.

The operating model

Local preparation

The data engine imports, profiles, indexes, scans, and stores working intelligence locally.

Explicit model choice

The user can use:

• local lookups and exact calculations;
• a cloud AI provider you configure with your own API key;
• your own local model server; or
• Claude Desktop, through the connector you turn on.

Review before disclosure

The user reviews scan coverage, findings, masking, and the organizational rules that apply to the data.

The product boundary

RangerIO reduces preparation work and provides safety controls. It does not make an otherwise prohibited disclosure permitted, replace legal review, or guarantee that every sensitive value will be detected.

End-to-end workflow

1. 1Select files
2. 2Check their structure
3. 3Import and scan for personal data
4. 4Review and correct findings
5. 5Ask, report, or mask
6. 6Export locally or use your AI provider

Step 1 — Select and inspect

For file imports, RangerIO performs a light structure scan before the full import:

• tables show their type, size, rows, and columns;
• files that cannot be read show an error before the import proceeds;
• you can remove a file before importing.

This structure scan is not the final PII scan.

Step 2 — Import and scan

The import creates the working source, reads its schema and structure, indexes it for search, and scans it for personal data. Imports run in the background, so you can keep working. A green tick on My Data means the file is ready for questions.

The Intelligence Dossier, a richer profile of the file, is not built during import. You build it when you need it from the file's details (see The Dossier).

Step 3 — Privacy coverage

Import completion and PII-scan completion are separate states. A file may be imported while its privacy scan is still pending or incomplete. Questions wait until the scan is done.

Step 4 — Review readiness and correct findings

Open the file's details and PII review after processing. Review structure, data-quality signals, scan coverage, and findings. Correct false positives or missed values and apply your decisions; RangerIO rescans the file to match. A refresh re-imports the file; a rescan refreshes detection and coverage.

Step 5 — Route the task

Local and external model paths

Important distinction: in-app cloud vs. MCP

For an in-app cloud call, RangerIO can mask outbound values and rehydrate supported response values locally.

For an external MCP host, public rehydration is blocked because an MCP tool result is sent back through that host's conversation context.

Current desktop builds

• macOS: Apple Silicon, arm64
• Windows: x64

Startup requirements

Actual working capacity depends on file size, file complexity, and other applications using memory.

Install on Windows

1. 1Download the current Windows x64 installer.
2. 2Run the installer. It downloads the application package while it installs.
3. 3Accept the license and choose the install folder.
4. 4On the last page, choose Run RangerIO. You can tick Delete the downloaded installation files to free about 3.4 GB; future updates then download the full package.
5. 5Complete the in-app setup.

Install on macOS

1. 1Download the current Apple Silicon disk image (.dmg).
2. 2Open the disk image.
3. 3Move RangerIO to Applications.
4. 4Start RangerIO.
5. 5Complete any macOS security prompt and the in-app setup.

First launch

RangerIO starts its local data engine, checks your computer, and asks you to connect an AI model. You can choose Skip for now and connect one later. On the last step you can also turn on the Claude Desktop connector.

Connect an AI model

Import, privacy review, and standard exports do not need an AI model. Questions on Explore, the Intelligence Dossier, and AI-assisted reports do. RangerIO does not include a model: you connect your own.

#### Connect a cloud provider

1. 1On the home screen, choose + Add Model Provider in the Active AI Model panel. (Or open Models from the ⋮ menu and choose Cloud / API.)
2. 2Choose the provider: OpenAI, Anthropic, Fireworks AI, or OpenAI-compatible. The API Endpoint is filled in for known providers.
3. 3Paste the API Key.
4. 4Choose the model from the list, or enter a model ID confirmed by your provider.
5. 5Enter a Profile name.
6. 6Choose Test Connection, then Save & Connect.
7. 7Confirm the model shows as active in the Active AI Model panel.

Use a large, current model. Small models can fail on long files with a “context too long” error. Provider credentials, charges, retention, and contractual terms belong to that provider.

#### Connect a local model server

1. 1Start the server and load at least one model in it.
2. 2Open Models from the ⋮ menu.
3. 3Choose Llama / vLLM / Network Model.
4. 4Under Local LLM servers, choose Connect beside the running server.
5. 5If no server is detected, choose Configure or enter a custom endpoint, then choose Test & Connect.
6. 6Confirm the server appears as Active under Saved Configurations.

#### Switch, edit, or remove a saved model

1. 1On the home screen, open Models and Providers.
2. 2Under Saved Models, choose Make Active on the model you want to use. Check now confirms it still answers.
3. 3Use Choose model to pick a different model from the same provider.
4. 4Use Edit to change the endpoint, model, name, or key, then choose Save Changes.
5. 5Use Remove only when the saved model is no longer needed.

Removing a saved model does not delete imported files.

Working-data location

RangerIO uses a local data directory under the signed-in user's home directory. The current engine uses RangerIO_local as its local working directory. Imported working data, indexes, metadata, registries, and audit information can be present there.

Updates

macOS: the app checks for updates and asks before downloading; choose Update & Restart to install. Windows: automatic updates are off. Choose Check for Updates in the ⋮ menu, then Open RangerIO Releases to download the latest signed installer.

Troubleshooting startup

If RangerIO cannot start a required component:

1. 1Read the message RangerIO shows.
2. 2Close memory-heavy applications if the system is low on RAM.
3. 3Retry the action.
4. 4If a model is needed, check that your AI provider answers: on the home screen, open Models and Providers and choose Check now.
5. 5If the issue persists, open the ⋮ menu and choose Generate Support Bundle (or Export Support Snapshot at the top of Settings).
6. 6Save the snapshot, check what it contains, and email it to info@rangerio.com.

Projects and project groups

A project organizes related files for an engagement or analysis. A project group organizes multiple projects.

#### Create a project

1. 1Open My Data.
2. 2Open Actions.
3. 3Choose New Project.
4. 4Enter a project name.
5. 5Choose Create.

#### Create a project group

1. 1Open My Data.
2. 2Open Actions.
3. 3Choose New Project Group.
4. 4Enter a Group Name, choose a color, and add projects to the group.
5. 5Choose Create.

Verified import formats

Tabular and structured

• CSV
• TSV
• XLS
• XLSX
• XLSM
• JSON
• JSONL
• Parquet

Documents and text

• PDF
• DOCX
• PPTX
• HTML
• TXT

Import one or more files

1. 1Open My Data.
2. 2Choose + Add Files. The Import Data dialog opens.
3. 3Choose Files and pick one or more files.
4. 4Check each file's type, size, rows, and columns. Remove anything picked by mistake with the ×.
5. 5Optionally open Advanced Options to set the Domain (default Auto-detect), a Reporting Profile, and Key Notes, for example “Q4 audit data; focus on vendor transactions above $10,000.” Most imports can leave these as they are.
6. 6Choose Start Import.
7. 7Follow progress in the import card or in Events in the header.
8. 8New files go to the Library project. To file them elsewhere, choose Move to project on the file, or select files and choose Move selected.

The Domain, Reporting Profile, and Key Notes fields guide later intelligence and reporting. They do not change the file's data.

Drag files onto My Data

My Data accepts dragged files and folders. They open the import dialog with the files already listed. You still review them and start the import.

Import a Cowork Folder

A Cowork Folder is a folder RangerIO watches. Its files are imported into their own project, and RangerIO checks the folder for new and changed files and asks you before applying anything. Subfolders are not imported.

1. 1Open My Data.
2. 2Choose + Add Files → Cowork Folder.
3. 3Choose the folder.
4. 4Select the File Types to include.
5. 5Choose Scan Directory.
6. 6Review the discovered files. Use Select All, Deselect All, or select individual files.
7. 7Choose Continue with N files.
8. 8Follow the import tasks. The files appear in a new project named Directory: followed by the folder name.

When files in the folder change, the project shows N changes. Open the project, review the list, and choose Review & Sync to apply the changes or Dismiss to ignore them.

During import

The task system reports current work and can show partial, interrupted, paused, failed, or completed states.

If an import is interrupted after a restart:

1. 1Look for the banner on My Data.
2. 2Choose Resume (or Continue preparation) to continue when resources permit.
3. 3Choose Dismiss to stop the unfinished work instead.
4. 4If Resume is refused because memory is low, close other applications and retry.

What appears after import

A file's row and details can show:

• file name and format;
• project placement;
• row or document information;
• a green tick when the file is ready for questions;
• privacy severity and scan coverage;
• data-quality and structure warnings;
• Intelligence Dossier status;
• Preview Data, Chat, Refresh, Move to project, and Delete actions.

The My Data toolbar also supports:

• search;
• Filter by status or severity, and Sort by;
• selecting one or more files;
• Use in Chat when two or more ready files are selected;
• Refresh selected and Move selected; and
• Delete.

Source status

Use these concepts carefully:

• Ready (green tick): the file is ready for questions.
• Needs attention: a warning, incomplete process, or uncertain structure requires review.
• Failed: the import did not complete. The message says why and what to do.
• PII scan required: the privacy scan has not completed.
• · unverified after a severity: PII findings may be real, but the scan did not finish fully.

A source can be analytically ready while its privacy coverage still needs attention.

Excel workbooks

A workbook is shown as one file with its sheets beneath it. Users should:

1. 1Open the workbook on My Data.
2. 2Review each sheet's rows, columns, quality, and PII information.
3. 3Ask about the specific sheet when a question must be limited to it.
4. 4Verify which file and sheet an answer or report used.

Refresh a source

1. 1Open My Data.
2. 2Select the file.
3. 3Choose Refresh on the file, or select files and choose Refresh selected.
4. 4Follow the task in Events.
5. 5Review structure and PII coverage again.

A refresh re-imports the file. If it has an Intelligence Dossier, rebuild it afterwards.

Delete a source

1. 1Select the file or files.
2. 2Choose Delete (for a workbook, Delete workbook).
3. 3Confirm the exact files.
4. 4Verify that the correct project remains.

Deletion removes RangerIO's working records for the selected files. It does not delete the original file on your computer.

What a dossier is

The Intelligence Dossier is RangerIO's reusable structured understanding of a file. It is not built during import: you build it when you need it. Reports need it, and questions and the Claude Desktop connector use it when it exists.

A dossier is not a second copy of the original document for human reading. It is a structured intelligence layer.

What can appear in a dossier

Depending on the source and completed processing:

• source name, format, size, and timestamps;
• row, column, page, paragraph, or document-structure information;
• schema and inferred types;
• per-column statistics such as missingness, uniqueness, common values, and numeric summaries where applicable;
• data-quality and parsing-confidence signals;
• detected entities;
• PII findings and scan coverage;
• searchable chunks and retrieval metadata;
• domain and reporting-profile context;
• workbook/sheet attribution; and
• provenance used by questions and reports.

Not every field is available for every format.

Build and open a dossier

1. 1Open My Data and select the file.
2. 2Wait until the privacy scan is complete and an AI model is connected.
3. 3Choose Build Intelligence Dossier in the file's details.
4. 4Wait for Intelligence Dossier · built followed by the time.
5. 5Choose Preview Dossier to read it.

Building a dossier uses your AI model. Personal details are replaced before anything is sent. If a dossier cannot be built, RangerIO says why (for example, the privacy scan is incomplete or no model is connected) and offers Retry.

You can also build dossiers during import: in Advanced Options, turn on Build Dossier Intelligence. It is off by default.

When the dossier changes

A dossier changes only when you build or rebuild it. If the file changes afterwards, RangerIO marks the dossier as being from an earlier version of the file; choose Rebuild to refresh it. Review its warnings again before relying on it or exporting.

Data-quality signals

RangerIO can surface structure confidence, missingness, anomalous values, normalization issues, and other data-quality context depending on the source and operation.

For analytical answers, caveats can be operation-aware. For example, a total over a materially incomplete column can carry a missingness warning.

A missing warning is not proof that the data is clean.

What the Privacy Scan does

RangerIO scans supported structured and document content for potential sensitive values. Detection can combine machine-validatable patterns, model-detected entities, prior user marks, and stored corrections.

Findings are reviewed in the PII review window. Each finding is labeled Critical or Non-critical.

Coverage and severity are different

• Severity describes the risk level of findings that exist.
• Coverage describes whether scanning completed sufficiently.
• PII scan required means the file has not been scanned. It is not a severity.
• · unverified after a severity means findings may exist but the scan is incomplete.

Zero findings on a file that says PII scan required does not mean “No PII.”

Review and correct findings

The same steps work for tables and documents.

1. 1Open My Data and select the file.
2. 2Choose Preview Data. The PII review window opens. (From a file preview, choose Privacy.)
3. 3Start with Needs Your Attention, which lists the PII types still to review, or open Browse detections to see every finding.
4. 4Use Next finding and Previous finding to move between findings, and Next match and Previous match to step through each occurrence.
5. 5Select a finding, cells, a column, or text, then choose PII, Not PII, or Change type.
6. 6If the scan missed something, select it and choose Mark Selection as PII, or use Add PII Values to enter a value.
7. 7Choose Review decisions, check the list, then Apply. Apply saves all your decisions together.
8. 8RangerIO rescans the file so protection matches your decisions. You can keep browsing while it finishes.

Show PII highlights and Hide PII highlights turn the marks on and off. A saved Not PII decision can be undone with Restore finding.

Correction scope

A decision applies to every matching occurrence. The decision card shows how many (Exact matches in this document, or in this sheet). Check the count before you apply.

Re-scan and export gate

Masked document export uses the last complete scan. After you apply decisions:

1. 1Wait for the rescan to finish.
2. 2If the scan did not finish, your decisions are still saved. Use the scan action RangerIO shows (Rescan or Run the scan again).
3. 3In the document preview, wait for Scanned [time] · export ready before using Export masked.

When the scan is out of date, the document preview shows Masked export paused until re-scan. Choose Re-scan now.

If the scan found nothing

A completed scan with no findings is not a guarantee. Review the file and use Mark Selection as PII for anything missed.

What automated detection cannot promise

• masking of health terms inside documents, which are not detected in this release;
• complete recall for every locale and identifier;
• perfect distinction between real and template data;
• correct classification of every organization, date, account reference, or short name;
• a legal conclusion about whether a value is personal information; or
• permission to disclose the resulting file.

Human review remains required.

Purpose

Safe Handoff is the set of controls used when an explicitly configured cloud or connector workflow receives RangerIO context.

In-app cloud-provider flow

1. 1Read privacy state. RangerIO checks the selected files' scan coverage and current registry.
2. 2Build task context. Only the context the question or report needs is assembled.
3. 3Replace detected values. Registered values are replaced with stable stand-ins.
4. 4Check outbound content. RangerIO checks the request before every provider call. If it cannot confirm the request is sanitized, the cloud call is blocked.
5. 5Send to your provider. You configured this provider and are responsible for its account and terms.
6. 6Receive the response. RangerIO puts the original values back locally.
7. 7Show what was sent. Under the answer, Cloud request lists what was replaced before sending, and View lineage opens the full record.

Stable stand-ins

Stable stand-ins let repeated values remain linkable during an engagement. This can preserve relationships such as repeated counterparties without sending the original detected value.

The mapping is maintained in a local registry scoped to the engagement/project workflow.

Fail-closed behavior

If RangerIO cannot validate a cloud-bound context, the expected outcome is a blocked cloud request with an explanation—not a silent downgrade.

The user can then:

• rescan the file;
• correct findings;
• select fewer or different files; or
• use your own local model server.

External MCP flow

An external MCP host is different:

1. 1The host asks RangerIO to run a tool.
2. 2RangerIO authorizes the tool and builds a response.
3. 3PII-aware scrubbing and egress controls apply.
4. 4The tool result returns to the external host.
5. 5That result becomes part of the host's conversation context.

Because of step 5, RangerIO's public MCP surface blocks rehydration. The external host receives sanitized or tokenized content.

What MCP tools may return

Depending on the tool and scope:

• project and source metadata;
• dossier summaries;
• schema and quality information;
• PII summaries;
• sanitized row previews;
• sanitized analytical query results; and
• tokenized exports when the user explicitly enables that permission.

Review requirement

Masked context can still expose structure, categories, counts, excerpts, or undetected values. The user must decide whether the task and external provider are permitted.

Start a data question

1. 1Open Explore.
2. 2Choose + (Select a data source). The Select files to query panel opens.
3. 3Search for and tick one or more ready files. They appear as chips above the question box.
4. 4Preview a selected file when needed.
5. 5Type the question.
6. 6Type @ to reference a column by name where useful.
7. 7Press Enter.
8. 8Follow progress, and choose Cancel if the question should stop.
9. 9Review the answer and the signals under it.

On My Data, you can also choose Chat beside a ready file, or select two or more files and choose Use in Chat.

How RangerIO routes a question

RangerIO uses multiple answer paths:

1. 1Direct lookup. Schema, counts, known fields, and precomputed facts can be returned without an AI model.
2. 2Exact local calculation. Totals, counts, and averages that map to a safe query are worked out exactly on your computer, using every row, and nothing is sent to your AI provider. When a question does not map cleanly, RangerIO passes it to your model instead of guessing.
3. 3Model-assisted answer. Interpretive questions use your model with context from the files. Smaller tables are sent in full; very large ones as a sample, and the answer says so.

The user does not need to choose the internal route.

Example direct questions

• “How many rows are in this source?”
• “Which columns are present?”
• “What is the inferred type of Invoice Date?”

Example analytical questions

• “Show the top 10 vendors by total amount.”
• “Count transactions by status.”
• “List records over $10,000.”
• “Compare missingness in Q3 and Q4.”

Example interpretive questions

• “Summarize the main patterns in this source.”
• “What should I review first?”
• “Explain the largest differences between these two files.”
• “Draft a cautious summary that includes the data-quality caveats.”

Multiple sources

When selecting multiple sources:

• confirm that each source is ready;
• ask a question that identifies the desired comparison or join;
• review source attribution;
• reduce the number of sources if coverage or model limits become an issue; and
• do not assume all cross-source relationships are valid without checking keys and source structure.

Evidence and honesty signals

An answer can display:

• the files used;
• a coverage notice when only part of a table or document was used;
• how the answer was produced, for example Answered using masked data;
• a grounding badge for checked figures, for example Numbers traced to your data;
• data-quality caveats;
• a request to clarify when the question is unclear;
• a warning when a response was shortened;
• Cloud request, listing what was replaced before sending, with View lineage; and
• Validate metrics, an optional review that asks the model to check the answer against your data. Its status, the model used, and the time taken show in the bar at the bottom.

If a cloud request is blocked for privacy

1. 1Read the message. It lists the blocked files.
2. 2Choose Re-scan now.
3. 3Wait for Scan complete, then ask again.
4. 4If the scan finds problems, review and correct them first.
5. 5If RangerIO warns before sending that a file is not ready, you can choose Ask without it.

If an answer is incomplete

• ask a narrower question;
• select fewer sources;
• name the exact field or operation;
• review missingness and caveats;
• refresh or re-scan the source;
• add a related source only when it is genuinely needed; or
• calculate consequential numbers independently.

Limits of external-client answers

RangerIO can validate and sanitize its tool output. It cannot control or validate new prose that an external AI host generates after receiving that tool output.

What the connector is

RangerIO includes a local MCP server that lets an MCP-compatible host request approved RangerIO tools. The server is off by default.

The guided setup is for Claude Desktop. Other MCP hosts may work but are not covered by this guide.

Enable the MCP server

1. 1Open Settings.
2. 2Under General, expand Claude Desktop Connector.
3. 3Turn on MCP server.
4. 4Wait for the local connector service to start.
5. 5Choose + Add Connector.

Configure Claude Desktop

1. 1Make sure Claude Desktop is installed.
2. 2If RangerIO says Claude Desktop is open, close it.
3. 3Read “We'll add RangerIO to your Claude Desktop connectors” and choose Connect.
4. 4Open Claude Desktop and pick RangerIO from the connector menu.
5. 5Return to RangerIO and confirm the status changes to Connected after activity.

Connector status

• Not configured: the connector has not been installed.
• Pending: configured and waiting for the first connection.
• Connected: RangerIO has observed connector activity.
• Idle: configured without current activity.
• Error: the sidecar or configuration needs attention.

Use Re-test to check the connector and Disconnect to remove the configured connection.

Trust settings

The MCP connector is conservative by default.

Allow tokenized export from stdio agents is off by default. When off, an agent cannot request a tokenized project bundle outside RangerIO's data directory.

When enabled:

• a trusted agent can request that export;
• detected PII is replaced with tokenized/synthetic equivalents;
• the resulting bundle still leaves RangerIO's data directory; and
• the setting should be disabled when it is no longer needed.

Disable the server

Turning the MCP server off stops the local connector service and unregisters the managed desktop connection. Use this when connector access is not required.

A. Export standard data

1. 1Open Report from the ⋮ menu.
2. 2Choose the file under Select Source.
3. 3Leave AI Assisted off.
4. 4Review the preview.
5. 5Choose a Format.
6. 6Choose an option under Select Masking Strategy.
7. 7Choose Export followed by the format, for example Export CSV.
8. 8Follow progress, then choose Download.
9. 9Inspect the exported content before sharing it.

Format availability varies by file; the formats shown for the selected file are the ones available.

B. Generate an AI-assisted report

1. 1Make sure each file has an Intelligence Dossier (see The Dossier).
2. 2Open Report.
3. 3Choose one or more ready files under Select Source.
4. 4Turn on AI Assisted.
5. 5Enter a Report Title.
6. 6Choose a Report Template, or keep Custom and pick the Sections.
7. 7Choose the Output Format: DOCX, PDF, or XLSX.
8. 8Under Report masking, choose Mask sensitive values or No masking — original values. This applies to the downloaded report; requests to your AI provider stay protected either way.
9. 9Add Discoveries when offered.
10. 10Choose Generate Report (for a custom report, Generate DOCX Report, Generate PDF Report, or Generate XLSX Report). The report runs in the background.
11. 11For a template report, review the sections and their variants, then choose Download for all variants or Export Report for the selected ones. A custom report opens a save dialog when it is ready.
12. 12Read any Report notes and sections that could not be generated.

A generated report must still be reviewed for factual and professional accuracy.

If a report cannot be generated

If a file's Intelligence Dossier is not ready, generation stops with a message that names the file and what to do, and offers Retry. If some sections could not be generated, the report says so and lists Report notes in the document.

C. Export a masked document from the preview

1. 1On Explore, select the document, turn on Preview, and choose Open full preview.
2. 2Complete the PII review and Apply your decisions (see Privacy Scan).
3. 3Wait for Scanned [time] · export ready.
4. 4Choose Export masked. RangerIO saves a masked DOCX copy.
5. 5Follow progress.
6. 6Open the exported copy and inspect every sensitive section before sharing it.

D. Export from PII Inventory

For supported sources:

1. 1Open PII Inventory from the ⋮ menu.
2. 2Expand the file.
3. 3Review findings by type.
4. 4Tick the PII types to mask.
5. 5Choose the Strategy: Partial, Redact, or Hash.
6. 6Choose Export Masked.
7. 7Review the resulting copy.

Masking strategies

Select Masking Strategy on the Report page offers these options when the selected file supports them. When the file has critical findings, Partial is selected by default.

• None — export as-is: does not mask detected PII. Do not use it for a sensitive file you will share.
• Partial: shows the first and last characters, for example J***n.
• Full: replaces all characters of the detected value.
• Redact: removes the detected value from the output.
• Tokenize: replaces detected values with stand-ins that cannot be reversed.

Availability and behavior depend on the source and export path.

Audit trail

RangerIO records local audit information for protected operations, including connector and export activity where supported. Audit records are technical evidence of what RangerIO recorded; they are not a regulatory certification.

Check a document and create a masked review copy

Ask a spreadsheet question and verify the answer

Compare two files or reporting periods

Prepare a draft report for professional review

Review a folder of engagement files

Use RangerIO from Claude Desktop

Individual professional workflow

1. 1Create a project for the engagement.
2. 2Import supported files.
3. 3Review structure warnings.
4. 4Review the privacy scan.
5. 5Correct findings and apply your decisions.
6. 6Ask local or model-assisted questions.
7. 7Verify important answers.
8. 8Export the dossier, report, or masked copy appropriate to the task.
9. 9Retain or delete the local project according to organizational policy.

Accounting and audit workflow

1. 1Create a project for the client and period.
2. 2Import ledgers, workbooks, and supporting documents in verified formats.
3. 3Review sheet structure and header confidence.
4. 4Inspect missingness, unusual values, and parsing warnings.
5. 5Review PII across the sources.
6. 6Ask precise questions such as totals, counts, status breakdowns, and exceptions.
7. 7Read grounding and data-quality caveats.
8. 8Verify material totals against the underlying rows.
9. 9Export data or a report for professional review.

Legal and HR workflow

1. 1Create a project for the matter or HR review.
2. 2Import supported DOCX, PDF, spreadsheet, or text sources.
3. 3Open each document with Preview Data.
4. 4Start with Needs Your Attention in PII review.
5. 5Step through each occurrence.
6. 6Choose PII, Not PII, or Change type, and mark missed values.
7. 7Apply, and wait for the rescan to finish.
8. 8Ask questions with a permitted AI provider or your own local model server.
9. 9Export a masked copy only after reviewing the export itself.

Healthcare and research workflow

1. 1Create a project for the authorized dataset or study workflow.
2. 2Import supported tabular exports and documents.
3. 3Review the source's structure and scan coverage.
4. 4Inspect names, dates, identifiers, locations, and other potential re-identification fields.
5. 5Mark specialized identifiers that the scan missed. Health terms inside documents are not detected in this release: mark them yourself or leave those files out.
6. 6Apply, and wait for the rescan to finish.
7. 7Where policy requires it, use exact totals and counts (worked out on your computer) or your own local model server.
8. 8If an external provider is permitted, review sanitized context and organizational rules first.
9. 9Verify all analytical results against the source.

Repeated engagement workflow

1. 1Start a new project for the new period or client.
2. 2Import the current files.
3. 3Confirm that source structure has not changed unexpectedly.
4. 4Review new scan findings rather than assuming prior classifications apply.
5. 5Ask repeatable questions using the same field names and operations.
6. 6Export the current output and record the review.

It is not a regulatory certification

RangerIO provides technical controls and review tools. It does not make an organization HIPAA-compliant, SOC 2-certified, PCI DSS-compliant, or legally authorized to disclose data.

It is not DLP or endpoint security

RangerIO does not monitor all filesystem, email, browser, or network activity and does not enforce organization-wide data-loss policy.

It is not a cloud warehouse

The desktop app is workstation-oriented. It is not a centralized multi-user warehouse or enterprise identity/access-control plane.

It is not a real-time streaming pipeline

RangerIO imports files and can watch a Cowork Folder for new and changed files, which you review before they are applied. It is not a continuous event-stream pipeline.

It does not import from databases or REST APIs

The desktop app imports files and folders. Database and REST API sources are part of RangerIO Enterprise licensing, with full support and centralized management, and are expected to begin shipping in Q3 2027. For more information on our Enterprise version, please contact info@rangerio.com.

It does not certify a source as free of PII

A complete scan can miss values or produce false positives. A user must review the source and output.

It does not guarantee correct AI output

Models can be incorrect, incomplete, or misleading. Grounding and caveat controls reduce risk but do not eliminate it.

It does not control an external AI host

RangerIO controls its own tool response and egress checks. The host's final prose, retention, account settings, and provider behavior remain outside RangerIO.

It does not include an AI model

Questions, the Intelligence Dossier, and AI-assisted reports use the cloud provider or local model server you connect.