How to use Prompt Lab

A quick walkthrough of every feature in the workbench — from writing your first prompt to running Compare across providers.

Write Core

The main workspace, under Create → Write. Write or paste a prompt, refine it with AI, and save it to your library.

Quick start

Write or paste your prompt

Type in the input area or paste from your clipboard. A word/character/token counter updates live in the top-right corner.

Choose an enhancement mode

Use the dropdown to the left of the Enhance button. Options: Balanced, Claude (XML-optimized), ChatGPT (system/user cues), Image Gen, Code Gen, Concise, or Detailed.

Click Enhance or press ⌘⏎

Your active provider rewrites the prompt. The enhanced version appears below with a quality score (Role / Task / Format / Constraints / Context).

Review and iterate

Click Show Diff to see word-level additions and removals. If the lint panel flags issues, use the Fix buttons for auto-correction.

Save to your library

Press ⌘S or click Save. Add a title, pick tags, assign a collection, and optionally write a change note describing what you adjusted.

Quality scoring

The editor automatically scores your prompt on five dimensions:

  • Role — detected if you include "you are," "act as," or similar persona cues
  • Task — action verbs like "write," "generate," "explain"
  • Format — output format specified (JSON, markdown, list, table, etc.)
  • Constraints — guardrails like "must," "avoid," "limit"
  • Context — sufficient background (80+ characters)

Score color: green (4-5/5), yellow (2-3/5), red (0-1/5).

Golden Responses

When you get an output you're happy with, click Pin Golden. This saves it as a benchmark. Future iterations will show a side-by-side comparison against your golden reference, so you can tell if quality is drifting.

Tip: The editor also shows Variants — alternative phrasings generated alongside the main enhancement. Click Use on any variant to promote it.

Workspaces

The header tabs switch between the Create views and the other workspaces:

  • Write — full-width prompt editing
  • Library — browse, inspect, and organize saved prompts
  • Compose — build a prompt from saved blocks
  • Dual Pane — Write and Library side by side (wide screens)
  • Evaluate — run history and, with Pro, Compare
  • Scratch — freeform notes

On small screens the same destinations appear in the bottom navigation bar.

Library

Your prompt archive. Search, tag, version, and organize everything you've saved.

Browsing & filtering

  • Search — matches title, text, tags, and metadata in real time
  • Views — All prompts, Recent, Frequently used, Favorites, Templates, Incomplete, and Recently Deleted (kept for 30 days)
  • Status — filter by Draft, Active, Deprecated, or Not set
  • Tags — click a tag chip to filter (Writing, Code, Research, Analysis, Creative, System, Role-play, Other)
  • Collections Pro — custom folders you create; click a collection name to filter
  • Sort — Newest, Oldest, Most used, A–Z, Z–A, Collection, or Manual (move up / move down buttons)

Per-entry actions

  • Select a card — opens the inspector with Details, Content, Variables, Variants, Notes, Tests, and Version history tabs
  • Editor / Open in Editor — sends the prompt back into Write
  • Copy, Favorite, Duplicate — quick actions on each card and in the inspector footer
  • Composer — inserts the prompt as a block in Compose
  • Checkboxes — select several prompts for bulk tag, favorite, duplicate, copy, export, or move to Recently Deleted

Version history

Every time you save changes to an existing prompt, the previous version is preserved. Open the inspector's Version history tab:

  • Pick any two versions in Compare from / Compare to for a word-level comparison
  • Click Restore as current to bring back an earlier version; the current content stays in history

Import & Export

Optional pack: Image Generation (34 prompts, JSON). Download it, then use Library Import to review and add the prompts. It is not installed automatically.

Already loaded the engineering starters? Download Prompt Engineering (12 templates, JSON) and review it in Library Import. Choose replacements explicitly to update existing copies; importing is never automatic.

In the Library toolbar (free on every plan):

  • Export — downloads your entire library as a .json file; Export selected in the bulk bar exports just the checked prompts
  • Import — loads prompts from a JSON file. Supports both Prompt Lab's native format and external schemas (fields like prompt, description, category are auto-mapped)

Compare Pro

Run 2–5 prompt or model variants side by side under Evaluate → Compare. Compare is part of Prompt Lab Pro on the hosted web app.

How to run a test

Open Evaluate → Compare

Click Evaluate in the header, then Compare. From the Library you can also choose Compare in the inspector footer.

Enter your variants

Fill in each variant's prompt and model. Use + Variant to add up to five.

Run the variants

All variants run concurrently. Results appear side by side; the diff view compares the baseline with another completed output.

Pick a winner

After reviewing the outputs, click Pick on the best variant to record the winner, or Promote to Golden to keep it as a benchmark. Runs are saved to Evaluate → History.

History & recent runs

Evaluate → History is free and lists every run with filters for model, provider, status, date, and verdict.

Tip: To test the same prompt across two different models (e.g., Claude vs. GPT-4), change your provider in Settings between runs, or use the Run A / Run B buttons individually after switching providers.

Compose

Build complex prompts from modular blocks — drag, drop, and reorder pieces from your library.

Three-panel layout

  • Library panel (left) — your saved prompts. Click + or drag to add a block to the canvas
  • Canvas (center) — your working composition. Drag blocks to reorder. Click X to remove
  • Preview (right) — live preview of the final composed prompt with markdown headers and section separators

On mobile, these collapse into Library | Canvas | Preview toggle tabs.

Output actions

  • Copy All — copies the entire composed prompt
  • Send to Editor — loads the composition into Write for refinement
  • Clear — removes all blocks from the canvas

Scratch

A freeform notepad for ideas, snippets, and working notes. Supports multiple named pads.

Managing pads

  • New — creates a new named pad (prompted for a name)
  • Rename — change the active pad's name
  • Delete — remove the active pad (disabled if only one remains)
  • Switch between pads using the sidebar list on the left

Toolbar

  • Date — inserts a formatted date separator (e.g., "── Fri, Mar 14, 2025 ──")
  • Download — exports the pad as a .txt file
  • Copy — copies all content
  • Clear — empties the pad (with confirmation)

Auto-save

Content saves automatically as you type. The status indicator in the bottom-right shows:

  • Saving... — debounced write in progress
  • Saved — last save succeeded (with timestamp)
  • Save failed — storage error (rare; usually means localStorage is full)

Template Variables

Write prompts once with placeholders, fill them differently each time you run.

User-defined variables

Wrap any placeholder in double braces: {{role}}, {{language}}, {{context}}. When you enhance, Prompt Lab detects all {{ }} patterns and shows a form to fill in values before sending.

Ghost variables (auto-resolve)

These built-in variables fill themselves at runtime — no input needed:

VariableResolves to
{{date}}Current date (YYYY-MM-DD)
{{time}}Current time (HH:MM:SS)
{{datetime}}Combined ISO format
{{timestamp}}Milliseconds since epoch
{{year}}Current year
{{clipboard}}Contents of your system clipboard
Example: Review this {{language}} code for security issues: {{clipboard}} — the language variable prompts you, and the clipboard content is injected automatically.

Settings

Configure providers and API keys. Click the gear icon in the header or press ⌘K and choose "Open Settings."

Providers

Select your active provider from the dropdown. Each provider has its own fields:

  • Anthropic — API key (sk-ant-...), model (default: claude-3-5-sonnet-20241022)
  • OpenAI — API key, model (default: gpt-4o)
  • Google Gemini — API key, model (default: gemini-2.0-flash)
  • OpenRouter — API key, model (default: anthropic/claude-sonnet-4.6)
  • Ollama — base URL (default: localhost:11434), model. Click Refresh Models to auto-detect loaded models

Testing your connection

Click Test Connection to validate your API key and model. A success or error message confirms the result. Keys are stored in localStorage and never leave your browser.

PII detection

If Prompt Lab detects sensitive content in your prompt (API keys, passwords, tokens), it shows a warning with three options:

  • Redact & Send — replaces sensitive data with placeholders
  • Send Anyway — proceeds without redaction
  • Cancel — aborts the operation

Keyboard Shortcuts

Press ? anywhere to see the full shortcut list, or ⌘K to open the command palette.

Global

ShortcutAction
⌘⏎Enhance prompt
⌘SSave prompt
⌘KOpen command palette
?Show keyboard shortcuts
EscClose modal / panel

Scratch

Scratch shortcuts are limited to local text actions. Browser-reserved shortcuts like ⌘T, ⌘W, ⌘[, and ⌘] are intentionally not overridden.

ShortcutAction
⌘EExport / download pad
⌘⇧DInsert date separator
⌘⇧CCopy all content
⌘⇧XClear pad

Command palette

Press ⌘K to open the palette. Type to filter, then select:

  • Enhance Prompt, Save Prompt, Clear Editor
  • Go to Write / Library / Compose / Evaluate / Scratch
  • Toggle Light / Dark mode
  • Export Library, Open Settings
  • Show Keyboard Shortcuts
↑ Back to top

Prompt Lab v1.8.0 — Home · Quick start · Provider setup · Import / export · Variables · Shortcuts · Privacy · Source