Programming assistant that answers coding questions, explains errors, reviews snippets, and suggests refactors. Supports Python, JavaScript/TypeScript, Go, Rust, Java, SQL, and most mainstream languages. Paste the code and describe the issue; responses include explanations, not just fixes. Prompt example 1: Write a small Python script that reads every csv file in a folder, merges them into one file and removes duplicate rows, with usage notes. Prompt example 2: Why does this JavaScript print undefined? Find the cause and fix it: function sum(a,b){ a+b; } console.log(sum(1,2)); Prompt example 3: Review this code, identify the security issue and show a safer version: query = "SELECT * FROM users WHERE name='" + name + "'" Prompt example 4: This Java code throws ConcurrentModificationException. Explain why and fix it: for (String s : list) { if (s.isEmpty()) list.remove(s); } Prompt example 5: This Python loop removes duplicates from 100,000 items with a list and takes minutes. Find the bottleneck and rewrite it: for x in data: if x not in result: result.append(x) Prompt example 6: Write pytest unit tests for this function covering normal, edge and invalid inputs: def discount(total, rate): return round(total * (1 - rate), 2) Prompt example 7: Rewrite this JavaScript callback with async/await and add error handling: fs.readFile('a.txt', (err, data) => { if (err) throw err; console.log(data); }); Prompt example 8: This C++ code leaks memory when it returns early on an error. Rewrite it with smart pointers and explain the difference: int* buf = new int[100]; if (!load(buf)) return; delete[] buf; Prompt example 9: Write a Bash script that compresses /var/www into /backup at 2 am every day, keeps only the last seven days of backups, and include the crontab entry. Prompt example 10: Rust reports borrow of moved value for: let s = String::from("hi"); let t = s; println!("{}", s); Explain ownership and fix it. Maximum context window (input and output combined): about 1,048,576 tokens. Maximum output per reply: about 65,536 tokens (model reasoning, if any, counts toward this limit). Output content formats: Text Pricing: Medium Continuous conversation: Supported AI can make mistakes. Please verify important information.