Run multi-step AI workflows on ChatGPT and Gemini — hands-free.
PromptChain is a free, open-source Chrome extension that queues your prompts and runs them one at a time. It types each instruction, hits send, waits for the full response, then moves to the next step automatically.
Perfect for research pipelines, content workflows, code reviews, tutoring sequences, and any task where you’d normally copy-paste prompts manually.
| Without PromptChain | With PromptChain |
|---|---|
| Copy prompt → paste → wait → repeat | Queue all steps once → walk away |
| Lose track of which step you’re on | Live progress overlay on the page |
| Same prompts for ChatGPT and Gemini in different tabs | Separate saved lists per platform |
| No signal when a long run finishes | Optional completion chime |
- Multi-step queue — Add as many instructions as you need
- Auto-submit & wait — Detects when the AI finishes generating before continuing
- ChatGPT & Gemini — Platform tabs with separate instruction sets
- Progress overlay — Floating panel on the chat page with step count and stop button
- Completion sound — Optional 3-note chime when all steps finish
- Configurable delay — Pause between steps (0–60 seconds)
- Persistent storage — Instructions and settings saved locally in your browser
- Stop anytime — Cancel mid-run from the popup or the on-page overlay
- Privacy-first — No accounts, no analytics, no data leaves your browser
| Platform | URL | Status |
|---|---|---|
| ChatGPT | chatgpt.com | Supported |
| ChatGPT (legacy) | chat.openai.com | Supported |
| Google Gemini | gemini.google.com | Supported |
Note: PromptChain is not affiliated with, endorsed by, or sponsored by OpenAI or Google.
-
Clone this repository
git clone https://github.com/neerajgrg/AiInstructions.git cd AiInstructions -
Open Chrome extensions
- Navigate to
chrome://extensions - Enable Developer mode (top-right toggle)
- Navigate to
-
Load the extension
- Click Load unpacked
- Select the project folder
-
Pin the extension (optional)
- Click the puzzle icon in Chrome’s toolbar → pin PromptChain
Coming soon.
- Open ChatGPT or Gemini in a browser tab and sign in.
- Click the PromptChain icon in your toolbar.
- Select the matching platform tab (ChatGPT or Gemini).
- Add your instructions — one prompt per step.
- Adjust settings if needed:
- Delay between steps — seconds to wait after each response (default: 1s)
- Play sound when done — completion chime (on by default)
- Click Run on ChatGPT or Run on Gemini.
- Watch the overlay on the chat page, or switch tabs — the run continues in the background.
Step 1: Summarize this article in 5 bullet points: [paste article]
Step 2: Turn the summary into a LinkedIn post under 200 words
Step 3: Suggest 5 hashtags for the post
Step 4: Rewrite in a more casual tone
PromptChain runs each step only after the previous response is fully complete.
flowchart LR
A[You queue steps] --> B[Extension fills prompt]
B --> C[Clicks Send]
C --> D[Waits for response]
D --> E{More steps?}
E -->|Yes| F[Delay if configured]
F --> B
E -->|No| G[Sound + Done overlay]
Under the hood:
- A content script runs on supported chat pages
- It finds the composer input and send button using resilient DOM selectors
- It monitors generation state (stop button / streaming indicators)
- The popup stores your instruction lists per platform in
chrome.storage.local
AiInstructions/
├── manifest.json # Extension manifest (MV3)
├── background.js # Service worker
├── content/
│ ├── content.js # Automation logic (ChatGPT + Gemini)
│ └── overlay.css # On-page progress overlay
├── popup/
│ ├── popup.html # Extension popup UI
│ ├── popup.css
│ └── popup.js
├── icons/ # Extension icons
└── test/ # Automated tests (Playwright + mock page)
- Node.js 18+
- Google Chrome or Chromium
Tests use Playwright with a local mock chat page (no API keys or login required).
npm install
npm testThis launches Chromium with the extension loaded and verifies the full two-step run flow.
- Go to
chrome://extensions - Click the reload icon on PromptChain
- Refresh any open ChatGPT/Gemini tabs
| Permission | Why it's needed |
|---|---|
storage |
Save your instruction lists and settings locally |
activeTab |
Interact with the current chat tab when you click Run |
scripting |
Inject the content script if it hasn't loaded yet |
Host access (chatgpt.com, gemini.google.com) |
Run automation only on supported AI chat pages |
PromptChain does not request access to all websites, <all_urls>, or your browsing history.
- Chrome Web Store release
- Claude.ai support
- Named instruction presets (save/load workflows)
-
{{previous_response}}variable in steps - Browser notification on completion
- Import / export instruction sets
- Start from step N
Contributions and feature requests are welcome — see Contributing.
Extension doesn't appear after loading
Make sure you selected the folder containing manifest.json, not a parent directory. Reload the extension from chrome://extensions.
"Open ChatGPT/Gemini in a tab first"
You need an open, logged-in tab on the matching platform. Switch to the correct tab in the popup before running.
Prompt not sending / stuck on a step
The AI site may have updated its UI. Try refreshing the page. If the issue persists, open an issue with your browser version and platform.
No completion sound
Ensure Play sound when done is checked and the tab isn't muted. Chrome may block audio on background tabs — the overlay will still show completion.
- Fork the repository
- Create a feature branch (
git checkout -b feature/my-feature) - Commit your changes (
git commit -m 'Add my feature') - Push to the branch (
git push origin feature/my-feature) - Open a Pull Request
Please keep PRs focused and include test updates when changing automation logic.
MIT License — free to use, modify, and distribute.
Built by neerajgrg
If PromptChain saves you time, consider starring the repo — it helps others discover it.