├── Basic ├── userinput.py ├── readme.md └── rules.md ├── .gitattributes └── readme.md /Basic/userinput.py: -------------------------------------------------------------------------------- 1 | user_input = input("prompt: ") -------------------------------------------------------------------------------- /.gitattributes: -------------------------------------------------------------------------------- 1 | # Auto detect text files and perform LF normalization 2 | * text=auto 3 | -------------------------------------------------------------------------------- /Basic/readme.md: -------------------------------------------------------------------------------- 1 | # 🔁 Interactive Task Loop with User Feedback 2 | 3 | This tool enables an AI-driven workflow where tasks are performed interactively in response to user input. After each task, the user is prompted for the next instruction. The loop continues until the user stops it manually or the maximum number of tool calls is reached. 4 | 5 | --- 6 | 7 | ## 🧠 How It Works 8 | 9 | 1. The AI performs its assigned task. 10 | 2. Upon completion, it runs a simple Python script to prompt the user for feedback or a new instruction. 11 | 3. Based on the user's input, the AI continues working. 12 | 4. The loop stops when: 13 | - The user manually stops, or 14 | - The tool call limit is reached. 15 | 16 | --- 17 | 18 | ## 📁 Files 19 | 20 | ### `userinput.py` (placed in the project root) 21 | 22 | ### `rules.md` ( copy the contents to your rules page) 23 | 24 | 25 | -------------------------------------------------------------------------------- /Basic/rules.md: -------------------------------------------------------------------------------- 1 | --- 2 | description: 3 | globs: 4 | alwaysApply: true 5 | --- 6 | 7 | ### ✅ Task: Interactive Task Loop with User Feedback 8 | 9 | 1. **Check if `userinput.py` exists** in the root directory. 10 | 11 | * If it doesn't exist, create it with the following content: 12 | 13 | ```python 14 | # userinput.py 15 | user_input = input("prompt: ") 16 | ``` 17 | 18 | 2. **Main Workflow**: 19 | 20 | * Perform your assigned tasks. 21 | 22 | * Run: 23 | 24 | ```bash 25 | python userinput.py 26 | ``` 27 | * The terminal should be opened in the chat window itself. 28 | 29 | * Read the user's input. 30 | 31 | * Based on the input, perform the next set of tasks. 32 | 33 | * Repeat the process. 34 | 35 | 3. **Exit Condition**: 36 | 37 | * If the user enters `"stop"` when prompted, exit the loop and terminate the process. 38 | -------------------------------------------------------------------------------- /readme.md: -------------------------------------------------------------------------------- 1 | # 🚀 10x-Tool-Calls 2 | 3 | **10x-Tool-Calls** is a simple rules setup designed for the **Cursor IDE**, **Windsurf**, or any other agent-based coding assistant that supports tool calls. It helps you get the **maximum value out of your monthly tool call allowance** by running your tasks in a loop with user input—without restarting the chat every time. 4 | 5 | Note : This only works with Agent Mode 6 | --- 7 | 8 | ## ✅ What It Does 9 | 10 | - After the AI completes a task, it runs a small Python script that asks: 11 | 12 | ``` 13 | 14 | prompt: 15 | 16 | ```` 17 | 18 | - You type your next instruction (e.g., `"add comments"`, `"refactor this"`, etc.) 19 | - The AI uses that input to continue working. 20 | - This loop repeats until: 21 | - You you manually stop, or 22 | - The session hits your **tool call limit**. 23 | 24 | --- 25 | 26 | ## 💡 Why This Matters 27 | 28 | Most AI coding tools (like Cursor) offer **500 monthly requests**, but each request can include **up to 25 tool calls**. Normally, even saying `"hi"` uses up a full request, wasting potential. 29 | 30 | With **10x-Tool-Calls**: 31 | - You start with one request. 32 | - Inside that session, you can give **multiple follow-ups**. 33 | - All follow-ups run within the same request using available tool calls. 34 | - This means you get **10x (or more)** actual work done using the same quota. 35 | 36 | --- 37 | 38 | ## ⚙️ How To Set It Up ( Basic Version ) 39 | 40 | ### 1. Copy `userinput.py` in your root directory 41 | 42 | This script is used to collect user input between tasks. 43 | 44 | ### 2. Copy `rules.md` into your IDE 45 | 46 | * Open your IDE's rule configuration (e.g., in Cursor: .cursorrules file or project rules - set to “always”) 47 | * Paste in the rules from `rules.md` (included in this repo) 48 | 49 | That’s it! The system is now set to run in an interactive loop. 50 | 51 | --- 52 | 53 | ## 🧪 Current Version 54 | 55 | * ✅ **Supports:** plain text input only 56 | * ❌ **Not yet supported:** image upload or file drops (coming soon!) 57 | 58 | A more advanced version is planned that will support: 59 | 60 | * Image uploads 61 | * File Drops 62 | 63 | --- 64 | 65 | ## ⚠️ Important Note 66 | 67 | **Only use this setup with tools that offer a tool call–based quota**, not token-based pricing (like OpenAI’s pay-per-token). This setup is designed to **maximize bundled tool calls**, not minimize token usage. 68 | 69 | --- 70 | 71 | ## 🧠 Example 72 | 73 | You have **500 monthly requests** and each request allows up to **25 tool calls**: 74 | 75 | * Normally: 76 | `You say "hi" → AI replies "hi" = 1 request used.` 77 | 78 | * With 10x-Tool-Calls: 79 | `You say "hi" → AI replies → runs prompt → you say "add comments" → AI continues... (up to 25 calls)` 80 | \= still **just 1 request used**, but you did much more. 81 | 82 | --- 83 | 84 | Get more done. Use fewer requests. Save your quota. 85 | 86 | **→ That’s the power of 10x-Tool-Calls.** 87 | 88 | ``` 89 | 90 | Let me know if you have any other requests 91 | ``` 92 | 93 | ## Support 94 | If you find this tool helpful, you can support the development by: 95 | - Buying me a coffee at https://ko-fi.com/perrypixel 96 | - UPI to kevinp@apl 97 | --------------------------------------------------------------------------------