💡 How to use these prompts:
Replace everything in [BRACKETS] with your specific details before sending.
Click Copy to copy any prompt to your clipboard instantly.
[USE DEEPSEEK V4 FLASH AT $0.09/M FOR VOLUME TASKS]
[Disable thinking mode for speed and cost]
Generate [NUMBER] [THING] following this exact pattern:
Pattern: [PASTE AN EXAMPLE OF WHAT YOU WANT]
Variations: [LIST THE VARIABLES THAT CHANGE BETWEEN EACH ONE]
Language: [LANGUAGE]
Output format: [HOW TO FORMAT THE OUTPUT — numbered, file-per-function, etc.]
Do not add explanation. Output the [THINGS] only, following the pattern exactly.
[USE DEEPSEEK V4 PRO WITH THINKING ON]
Analyse the following code deeply before answering my question. Think through what this code does, what assumptions it makes, and what edge cases it might fail on — before you respond.
My question: [YOUR SPECIFIC QUESTION ABOUT THE CODE]
Context: [DESCRIBE WHAT THIS CODE IS SUPPOSED TO DO]
Code: [PASTE CODE]
Use extended reasoning to analyse thoroughly, then give me a direct answer to my question.
[USE WITH SELF-HOSTED V4 PRO — all context stays on your infrastructure]
I am pasting my full codebase below. Read it completely before answering. I will ask multiple questions about it.
Questions:
1. [QUESTION 1 — e.g. where is the authentication handled?]
2. [QUESTION 2 — e.g. what happens when a database write fails?]
3. [QUESTION 3 — e.g. which functions have no error handling?]
Answer each question with specific file and line references. Do not guess — only answer from what you can see in the code.
Codebase: [PASTE CODEBASE]
Write a complete data pipeline for the following specification. Include error handling, logging, and retry logic.
Pipeline spec: [DESCRIBE WHAT DATA GOES IN, WHAT TRANSFORMATIONS HAPPEN, WHAT COMES OUT]
Source: [WHERE DATA COMES FROM — database, API, file, stream]
Destination: [WHERE PROCESSED DATA GOES]
Language: [PYTHON / NODE / GO / OTHER]
Failure handling: [WHAT TO DO WHEN A STEP FAILS — retry, skip, dead-letter queue]
Logging: [WHAT TO LOG AND WHERE]
Write the complete pipeline code. No stubs.
Write a complete bash script for the following automation task.
Task: [DESCRIBE WHAT THE SCRIPT SHOULD DO]
OS: [LINUX / MAC / BOTH]
Inputs: [WHAT ARGUMENTS OR VARIABLES THE SCRIPT TAKES]
Error handling: [WHAT SHOULD HAPPEN WHEN COMMANDS FAIL]
Output: [WHAT THE SCRIPT SHOULD PRINT OR PRODUCE]
Include:
- Shebang line
- set -euo pipefail (fail fast on errors)
- Usage/help message if called with wrong arguments
- Comments on any non-obvious step
Optimise the following SQL query. Show me the original, the optimised version, and explain exactly what you changed and why.
Database: [POSTGRESQL / MYSQL / MSSQL / SQLITE]
Current query performance: [DESCRIBE — slow on X rows, times out after Y seconds, etc.]
Table sizes: [APPROXIMATE ROWS IN EACH TABLE]
Indexes available: [LIST EXISTING INDEXES OR "UNKNOWN"]
Query: [PASTE QUERY]
After the optimised query, give me:
1. What specific changes you made and why each one improves performance
2. Any index you should create that does not exist yet
3. Whether the query logic itself can be simplified