Documentation

KRINO Dokim v1.5.0

What is KRINO Dokim?

KRINO Dokim is a desktop application that converts user stories and acceptance criteria into structured BDD (Behavior-Driven Development) test cases in Gherkin format. It connects to your choice of AI provider (cloud or local) and generates test cases you can export directly into your QA workflow.

Installation

Windows

  1. Download krino-dokim-windows-setup.exe from the download page.
  2. Run the installer and follow the prompts.
  3. Launch KRINO Dokim from the Start menu.

macOS

  1. Download krino-dokim-macos.dmg from the download page.
  2. Open the file and drag KRINO Dokim to Applications.
  3. Open from Launchpad.

Linux

  1. Download krino-dokim-linux.deb from the download page.
  2. Install: sudo dpkg -i krino-dokim-linux.deb
  3. Launch from your applications menu.

On first launch, sign in or start your free trial.

Quick Start

  1. Choose your AI provider. Select a provider from the dropdown (OpenAI, Anthropic, Google Gemini, Ollama, or LM Studio) and enter your API key if required.
  2. Enter a user story. Fill in the Title and Acceptance Criteria fields. User, Goal, and Reason are optional but improve test case relevance.
  3. Generate. Click Generate Test Cases (or press Ctrl+Enter). Test cases appear as formatted BDD cards within seconds.
  4. Export. Download your results as JSON, Excel, CSV, Gherkin, Jira Xray, Zephyr Scale, or TestRail.

Single Story Mode

Use single story mode for one user story at a time. The form has two required fields and three optional fields:

  • Title (required): A short name for the story. e.g. "User Login with 2FA"
  • Acceptance Criteria (required): The conditions the feature must satisfy. List each criterion on a new line.
  • User (optional): Who is performing the action. e.g. "Security-conscious user"
  • Goal (optional): What they want to do. e.g. "Log in using my password and a one-time code"
  • Reason (optional): Why they need it. e.g. "My account remains secure even if my password is compromised"

User, Goal, and Reason together form the "As a / I want to / So that" BDD user story statement. Providing them gives the AI additional context and produces more targeted test cases.

Specification Linting

Specification Linting analyzes your acceptance criteria before generation and flags issues that tend to produce weak test cases: vague language, untestable conditions, missing actors, and ambiguous outcomes.

Click Analyze in the single story form to run a lint check. Findings appear in a collapsible panel below your acceptance criteria. Each finding includes a suggested rewrite you can apply directly with Apply. Use Fix All to apply all suggestions at once, or dismiss findings you disagree with.

Linting is advisory and never blocks generation.

Available in single story mode.

Output Critique

Output Critique runs a second AI pass over your generated test cases and flags gaps in coverage: missing negative scenarios, redundant scenarios, vague expected results, and edge cases not addressed by the original criteria.

Click Review Output after generation to run a critique. Findings are read-only. Use them to decide whether to refine your acceptance criteria and regenerate, or accept the output as-is.

Available in single story mode.

Batch Processing

Batch mode processes multiple user stories from a single Excel (.xlsx) or CSV (.csv) file. This is the fastest way to generate test cases for an entire sprint or backlog.

Required columns

  • Title: Story name
  • Acceptance_Criteria: Acceptance criteria text

Optional columns

  • Issue_ID: e.g. JIRA-123. Auto-generated if blank.
  • User: Who is performing the action
  • Goal: What they want to do
  • Reason: Why they need it

Column names are case-sensitive. Download the sample template from the batch upload panel for a ready-to-fill file.

Batch workflow

  1. Switch to the Batch tab (Ctrl+2)
  2. Upload your file
  3. Review the preview: sort any column, filter rows by keyword, and deselect rows you want to exclude. Confirm row count and column mapping before proceeding.
  4. Click Start Batch. Processing runs story by story with a live progress indicator.
  5. Use Pause/Resume to hold processing mid-batch. Use Cancel to stop early and keep results generated so far.
  6. Export the full batch when complete.

Batch file preview

After uploading a file, the preview table lets you inspect and refine your batch before processing:

  • Sort: Click any column header to sort ascending, descending, or reset to original order.
  • Filter: Type in the search box to filter rows across all columns (case-insensitive).
  • Row selection: Use checkboxes to deselect rows you want to exclude. Deselected rows are skipped during processing without editing the source file.

Preview is capped at 500 rows. Files with more rows display a truncation notice showing the actual total.

AI Providers

KRINO Dokim connects to your choice of AI provider. Local providers keep all data on your machine. Cloud providers require an API key, which is stored locally and never sent to KRINO Systems.

Ollama (local, free)

  1. Download and install Ollama.
  2. Pull a model: ollama pull llama3
  3. Open KRINO Dokim. Ollama is auto-detected on startup.

LM Studio (local, free)

  1. Download and install LM Studio.
  2. Load a model, then go to the Developer tab and click Start Server.
  3. In KRINO Dokim, select LM Studio. If using a non-default port, update the server URL in Settings → AI Provider.

OpenAI

  1. Get an API key at platform.openai.com/api-keys.
  2. In KRINO Dokim, select OpenAI and paste your key into Settings → AI Provider.

Anthropic

  1. Get an API key at console.anthropic.com.
  2. In KRINO Dokim, select Anthropic and paste your key into Settings → AI Provider.

Google Gemini

  1. Get an API key at aistudio.google.com.
  2. In KRINO Dokim, select Google Gemini and paste your key into Settings → AI Provider.

Coverage Levels

Coverage level controls how many test cases the AI generates per story.

  • Quick: 3–5 scenarios: Happy path and one critical negative case. Good for rapid prototyping or stories with narrow scope.
  • Standard: 8–12 scenarios (Recommended): Happy path, key negative cases, boundary conditions, and edge cases. Suitable for most production stories.
  • Exhaustive: 15–20+ scenarios: All positive, negative, boundary, and edge scenarios. Use for high-risk or complex stories where maximum coverage is required.

Reference Examples

Reference Examples let you calibrate output style by uploading a .txt or .feature file with 1–3 of your own story/Gherkin pairs before generation. The AI mirrors your team's step naming, tag conventions, and scenario structure when generating new test cases.

To use Reference Examples:

  1. Enable Match My Team's Style in the single story or batch form to reveal the upload area.
  2. Upload a file containing up to 3 examples (use ## Example 1, ## Example 2 headings to separate blocks). Maximum 4,000 characters after stripping comments.
  3. Generate as usual. The AI applies your conventions to the output.

Download the sample template from the panel to see the expected format. Examples persist through batch checkpoint save and restore.

Available in single story and batch modes.

Export Formats

  • JSON: Structured output. Useful for importing into other tools or custom pipelines.
  • Excel (.xlsx): One row per test case. Suitable for test management tools that accept spreadsheet import.
  • CSV: Plain text, comma-separated. Compatible with any spreadsheet application.
  • Gherkin (.feature): Cucumber, SpecFlow, and Behave compatible. One scenario per test case.
  • Jira Xray (JSON): Import directly into Jira Xray test management.
  • Zephyr Scale (JSON): Import directly into Zephyr Scale test management.
  • TestRail (CSV): Bulk import into TestRail via CSV upload.

All formats include the AI reasoning alongside the test steps. Exports are available for individual stories and full batch results.

Keyboard Shortcuts

Shortcut Action
Ctrl+EnterSubmit / Generate
Ctrl+1Switch to Single Story tab
Ctrl+2Switch to Batch tab
Ctrl+,Open Settings
SpacePause / Resume batch processing
EscClose open panel
Esc × 2Reset app (double-tap)
?Show keyboard shortcuts

Subscription & Activation

KRINO Dokim requires an active subscription to generate and export test cases.

  • Free trial: 7 days, full access. A card is required at sign-up but not charged until the trial ends. Cancel any time during the trial at no cost.
  • Pro Monthly: $10/month. Cancel any time.
  • Pro Annual: $100/year. Best value.

Manage your subscription, download invoices, or cancel in Settings → Subscription.

Contact & Support

Questions not covered here? Email support@krinosystems.com.