Turn rough issue descriptions into structured, ready-to-run AI agent prompts — in seconds.
Features
- Structured prompts — transforms a vague issue into a complete prompt with Context, Task, Current State, Expected Outcome, Constraints, and Acceptance Criteria
- 4 provider options — works free with Ollama locally, or use Groq, Gemini, or Anthropic via API
- Model picker — switch providers and models directly from the panel header
- Streaming output — see the prompt generated word by word
- Copy to clipboard — one click to copy the generated prompt
Providers
| Provider | Free? | Setup |
|---|---|---|
| Ollama | ✅ Fully free | Install Ollama locally |
| Groq | ✅ Free tier | API key from console.groq.com |
| Gemini | ✅ Free tier | API key from aistudio.google.com |
| Anthropic | 💳 Paid | API key from console.anthropic.com |
Usage
- Open the command palette (
Ctrl+Shift+P) and run Promptly: Generate AI Agent Prompt - Or right-click in the editor and select Promptly: Generate AI Agent Prompt
- Describe your issue in the text area
- Optionally add tech stack and extra details
- Click Generate Prompt
- Copy the result and paste it into your AI agent
Configuration
Go to Settings → Promptly to configure:
- Provider — choose Ollama, Groq, Gemini, or Anthropic
- API Keys — add keys for cloud providers
- Models — select the model per provider
- Ollama URL — change if Ollama runs on a non-default port
Requirements
- For Ollama: Ollama running locally with at least one model pulled (e.g.
ollama pull llama3.2) - For cloud providers: a valid API key for the selected provider