Convert raw human intent into structured prompts, optionally enhanced using Google Gemini.
Project description
📌 intent2prompt
intent2prompt converts natural, incomplete user instructions into structured, context-aware prompts that are ready for any LLM.
It understands the user’s intention, identifies the task type, and transforms vague commands into well-defined prompts.
Optionally, it can refine and improve the output using Google Gemini.
🚀 Why intent2prompt?
AI responses are only as good as the prompts they receive. Users often type:
email boss leave tomorrow
sql sales last month region west
ppt on ai future
These are not prompts — they are intentions.
This library converts those intentions into prompts that models actually understand.
✨ Core Capabilities
✔️ Detects the nature of the task automatically
(email, SQL query, PPT outline, explanation, summary, or generic instructions)
✔️ Generates structured prompts including:
- Tone
- Output format
- Expected response style
- Additional constraints
✔️ Provides optional Gemini enhancement for:
- Better clarity
- Professional tone
- Domain-aware restructuring
- Expandable context
✔️ Works with one-line API calls:
convert() for offline use, enhance() for Gemini integration
📦 Installation
pip install intent2prompt
🛠️ Usage
1️⃣ Convert intent into a structured prompt (offline)
from intent2prompt import convert
prompt = convert("email boss leave for 2 days", tone="formal")
print(prompt)
2️⃣ Enhance the prompt using Gemini
from intent2prompt import enhance, set_gemini_key
set_gemini_key("YOUR_GEMINI_API_KEY")
print(enhance("email boss leave for 2 days", tone="formal"))
🧪 Example
Input
email boss leave for 2 days
convert() output
You are an AI assistant.
Use a formal tone. Write a complete email with greeting, body, and closing.
User intention:
"email boss leave for 2 days"
Generate a complete response that satisfies this intention.
enhance() output (Gemini)
Draft a professional email requesting a two-day leave from your manager.
Specify dates, provide a brief reason, and include a polite closing line.
🔍 Where to Use intent2prompt
- AI chatbots and assistants
- Customer support systems
- Email automation workflows
- HR or employee request assistants
- SQL prompt builders for BI tools
- Documentation and training bots
- Agentic AI pipelines where user input is raw
If your users type intent, this library gives you the prompt.
🧭 Future Scope
The library is designed with expansion in mind:
🚧 v0.2.0 planned features
- Support for OpenAI, Claude, and Perplexity models
- Domain-specific prompt packs:
- HR communication
- Finance & accounting workflows
- Power BI reporting prompts
- Restaurant & delivery ecosystem prompts
- Multi-language prompt generation
- Automatic tone detection
- Command Line Interface:
i2p "email boss leave"
🚀 Long-term vision
Make intent2prompt the foundational layer for every agentic AI system —
where users simply express intent, and the system generates the best possible prompt.
👤 Author
Rajdeep Rao
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file intent2prompt-0.1.1.tar.gz.
File metadata
- Download URL: intent2prompt-0.1.1.tar.gz
- Upload date:
- Size: 4.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b60ed8073ae30d8a5b6bc72ad036412f51cea46d25732699ab993447b87efe56
|
|
| MD5 |
64ab7da39910c51a7d5b84312e28e57a
|
|
| BLAKE2b-256 |
e76ab9b8d2ec54a445a4adb1d4a1cc18cfbb4e364153066a6e63983d1904f300
|
File details
Details for the file intent2prompt-0.1.1-py3-none-any.whl.
File metadata
- Download URL: intent2prompt-0.1.1-py3-none-any.whl
- Upload date:
- Size: 4.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e7ad81cadf2596d15a54dd8e862bf9886f2a07422d3cfc174cf4f1aa6b6aa3b8
|
|
| MD5 |
e9f3486e58b148f818d0f46ddfb0d2b9
|
|
| BLAKE2b-256 |
92fb88f18cbd53995a2f00f5fe0870642c33dd571f85b20893b41b1f362e24b2
|