Documentation

Everything you need to get the most out of PromptForge — from installation to advanced settings.

Installation

PromptForge is currently in beta and not yet listed on the Chrome Web Store. Download the beta build below and install it in under two minutes.

1. Download the zip above and unzip it to a permanent location — Chrome loads the extension from this folder each time
2. Open Chrome and go to chrome://extensions
3. Enable Developer mode (toggle in the top right)
4. Click Load unpacked and select the promptforge folder you just unzipped
5. The ⚡ icon appears in your toolbar — pin it
6. Open ChatGPT or Claude, type a prompt, and press Enter to see PromptForge in action

How it works

PromptForge attaches a keyboard listener to the page in capture phase. When you press Enter in a supported chat input, it intercepts the event before the site processes it, opens the enhancement modal, and waits for you to build and confirm your prompt. When you click Enhance & Send, it injects the enhanced prompt into the input field and submits it.

Your original prompt is always shown in the modal so you never lose what you typed. All fields in every framework are optional — leave anything blank and it simply won't appear in the assembled output.

Supported sites

Frameworks — Simple

Simple frameworks are best for everyday, short prompts. The auto-classifier defaults to this category for prompts below the word count threshold (default: 30 words).

RTF Simple

Role / Task / Format — the default framework for quick, everyday requests.

When to use: Short prompts under 30 words where you just need a clear role, task, and output format.

• Role
• Task
• Format

Auto-trigger signal: Fallback for all prompts below the word count threshold that don't match a more specific framework.

RACE Simple

Role / Action / Context / Expectation — adds context and a stated outcome to the basic RTF structure.

When to use: Short prompts where adding context and an expected outcome improves the response.

• Role
• Action
• Context
• Expectation

Auto-trigger signal: Keywords suggesting context or expected outcome (e.g. "given that", "so that", "in order to").

PACT Simple

Purpose / Action / Context / Topic — focuses on intent and topic scope.

When to use: When you have a clear purpose and topic but need to frame the action within a context.

• Purpose
• Action
• Context
• Topic

Auto-trigger signal: Keywords indicating clear intent or purpose (e.g. "I need to", "help me", "I want to").

TAG Simple

Task / Action / Goal — an ultra-compact three-field structure for rapid prompts.

When to use: Very short prompts where only the task, action verb, and end goal matter.

• Task
• Action
• Goal

Auto-trigger signal: Very short prompts (under ~10 words) with an action verb and a clear end goal.

Frameworks — Scenario

Scenario frameworks provide rich context framing for longer, more nuanced prompts. Auto-triggered for complex prompts above the word count threshold.

COAST Scenario

Context / Objective / Actions / Scenario / Task — situational framing with explicit scenario setting.

When to use: Prompts that benefit from a described scenario or situation before stating the task.

• Context
• Objective
• Actions
• Scenario
• Task

Auto-trigger signal: Keywords describing a situation or setting (e.g. "imagine", "scenario", "given the situation").

TRACE Scenario

Task / Role / Audience / Context / Expectation — audience-aware prompting.

When to use: When the audience or reader of the output matters as much as the content itself.

• Task
• Role
• Audience
• Context
• Expectation

Auto-trigger signal: Keywords referencing an audience or reader (e.g. "for my team", "aimed at", "suitable for", "audience").

RISEN Scenario

Role / Instructions / Steps / End goal / Narrowing — step-by-step workflow prompts.

When to use: When you need the AI to follow a sequence of steps to reach a defined end goal.

• Role
• Instructions
• Steps
• End goal
• Narrowing

Auto-trigger signal: Keywords implying sequential steps or a workflow (e.g. "step by step", "walk me through", "process", "procedure").

COSTAR Scenario

Context / Objective / Style / Tone / Audience / Response — full narrative and stylistic control.

When to use: Content creation prompts where style, tone, and audience all need to be specified.

• Context
• Objective
• Style
• Tone
• Audience
• Response format

Auto-trigger signal: Keywords related to writing style or tone (e.g. "formal", "casual", "professional tone", "write in the style of").

Frameworks — Complex

Complex frameworks use advanced reasoning techniques. Triggered for prompts that explicitly request structured reasoning or examples.

CoT — Chain of Thought Complex

Chain of Thought — instructs the AI to reason step-by-step before arriving at an answer.

When to use: Logic, math, analysis, or multi-step problems where showing the reasoning matters.

• Problem / Question
• Context
• Reasoning instruction (auto-appended: "think step by step")

Auto-trigger signal: Keywords implying reasoning or analysis (e.g. "reason through", "think step by step", "explain your logic", "analyse").

Few-Shot Complex

Few-Shot prompting — provide input/output example pairs so the AI learns the pattern from demonstrations.

When to use: When you need the AI to match a specific format or behaviour you can demonstrate with examples.

• Task description
• Example 1 (input → output)
• Example 2 (input → output)
• Example 3 (input → output)
• Your actual input

Auto-trigger signal: Keywords suggesting examples or patterns (e.g. "for example", "like this", "same format as", "here's an example").

Frameworks — Goal + Technique

Objective-driven frameworks that structure a prompt around a measurable goal.

SMART Goal

Specific / Measurable / Achievable / Relevant / Time-bound — based on PMI prompt formula documentation.

When to use: Goal-oriented prompts with a deadline, metric, or defined outcome.

• Specific
• Measurable
• Achievable
• Relevant
• Time-bound

Auto-trigger signal: Deadline or goal keywords — e.g. "by end of quarter", "OKR", "KPI", "goal", "timeline", "deadline", "target".

SMART works in guide mode — it assembles a structured prompt in the preview pane that you can edit before sending.

Auto-classifier

The classifier runs on every intercepted prompt and selects a framework based on signals checked in priority order:

1. Pinned default — if you've set a default framework in settings (not Auto), it always wins
2. SMART keywords — deadline, goal, timeline, OKR, KPI, "by end of quarter" etc. → SMART
3. Complex keywords — "step by step", "reason through", "for example" etc. → CoT or Few-Shot
4. Scenario keywords — audience, tone, style, scenario, workflow etc. → COAST / TRACE / RISEN / COSTAR
5. Word count — at or above threshold → scenario or CREATE-family framework
6. Fallback — everything else → RTF

A reason line appears below the framework badge in the modal explaining why the classifier chose that framework. You can always override using the framework toggle row.

Modal overview

The modal slides up from the bottom of the page when you press Enter. It contains:

• Header — logo, framework badge, Send as-is button, close button
• Framework toggle — switch between enabled frameworks manually
• Classifier reason — explains why this framework was chosen
• Your prompt — read-only display of what you typed
• Framework form — pre-filled fields, all optional
• Quality checklist — collapsible 8-point self-check
• Preview pane — assembled prompt, editable before sending
• Footer — Preview, AI Assist, Save, Download, Enhance & Send

Quality checklist

A collapsible section with 8 yes/no checkboxes based on PMI effective prompt characteristics. Checkboxes auto-tick as you fill in fields — you don't need to check them manually. As items are ticked, a colour bar updates from red → amber → green. The current score is shown in the collapsed header so you can see your quality rating at a glance without expanding the section.

• Is the prompt specific and clear?
• Does it provide enough context without irrelevant detail?
• Does it include a role or character?
• Is the tone appropriate for the desired output?
• Are examples included where helpful?
• Is the language free from jargon or ambiguity?
• Is the output format or structure specified?
• Have you included a reliability/validation check?

AI Assist

The AI Assist button in the modal footer assembles your current prompt and wraps it in a meta-prompt:

"Review the following prompt and suggest improvements for clarity, specificity, and structure. Return an improved version."

The full meta-prompt is copied to your clipboard. Paste it into ChatGPT or Claude to get AI feedback on your prompt. No API calls are made from the extension itself.

Preview & edit

The preview pane updates in real time as you fill in fields — you don't need to click a button to see the assembled prompt. The pane is a live textarea, so you can edit the assembled text directly before sending. What you see in the preview is exactly what gets sent.

Secondary framework suggestion (auto-advise)

After the classifier selects a framework, it may also suggest a secondary framework that could suit your prompt. This appears as a dismissible notice below the reason line — for example: "COSTAR may also work well for this prompt."

• To dismiss: Click the × on the suggestion chip.
• To switch: Click the suggested framework name to switch to it immediately.
• To disable: Turn off Show secondary suggestion in Settings.

Framework tour

New to PromptForge? The framework tour walks you through all 12 frameworks with example prompts and use-case guidance. To launch the tour:

• Click the ? button in the modal header to open the help-choose panel, then click Take the tour.
• Or visit the tour page directly.

The tour can be restarted at any time from the ? panel.

More / Less framework toggle

The framework toggle row in the modal shows 4 primary frameworks by default (RTF, CREATE, SMART, and one context-appropriate pick). Click More to expand the row and show all 12 enabled frameworks. Click Less to collapse it again. Frameworks you've disabled in Settings are hidden from the toggle row entirely.

Settings reference

Access full settings via the ⚙ icon in the toolbar popup, or right-click the extension icon → Options.

Keyboard shortcuts

To change the Alt+Shift+P shortcut, go to chrome://extensions/shortcuts and find PromptForge.

Prompt history

Every prompt enhanced via Enhance & Send is automatically saved. Access history from the toolbar popup → Saved Prompts, or via the Options page → Saved Prompts tab.

Per-entry actions: Copy the enhanced prompt, Download as a .md file, or Delete.

Bulk actions: Export all as JSON (for backup), Clear all. History is capped at 500 entries.

Privacy & data

PromptForge is designed to be fully private:

• All data is stored in chrome.storage.local — sandboxed to this extension on this machine only
• No external API calls from the extension
• No telemetry, analytics, or usage tracking
• No accounts, no sign-in, no cloud sync
• The only network activity on supported sites is your own conversation with ChatGPT or Claude

Permissions used: storage, activeTab, scripting — the minimum required to function.