#Claude

Stop using the wrong AI tools…👇 Writing → Claude Brainstorming → ChatGPT Research → Perplexity Image Gen → Nano Banana Video Gen → Higgsfield Apps & Websites → Claude Code SEO → Cytd.ai Meeting Assistant → Supernormal Learning → NotebookLM Automating Tasks → Manus Save this so you never forget. Comment “tools” and I’ll send you a full breakdown of the 25 best AI tools I use every week. Follow for more AI. #ai #aitools #chatgpt #claude #bestaitools2026

Free Claude Code 🤯 Comment “CLAUDE” and get the full setup guide in your DM. [Gemma 4, Ollama, Claude Code, AI Coding, Claude] #Claude #CodingAI #Ollama #Gemma4 #AIDev

We asked 5 AI systems who should be sacrificed. ChatGPT threw shade. Grok was Grok. Gemini went brutal. And then there was Claude. RESPECT🫡 #ai #chatgpt #claude #grok #gemini

The PEAK setup. What would you add? #ai #claude #startup #motivation #yc

READ CAPTION 👇 Comment “Prompt” and follow and I’ll send it to you. This was so fun to play with! #claude #vibecoding #ai

comment “claude” and i’ll send these resources!! #computerscience #csmajor #internship #claude #claudecode

The Prompt - “You are operating as a senior-level AI collaborator, not a generic assistant. Before responding to anything, internalize the following operating parameters for this session. Treat me as a professional in my domain. Calibrate your language, depth, and assumptions accordingly. Do not over-explain fundamentals unless I ask. Skip preamble, filler affirmations (“Great question!”), and unnecessary caveats. - Lead with the answer. Context and reasoning follow, never precede. - Default to prose over bullet points unless the information is genuinely list-shaped. - Never use em dashes. Avoid passive voice. No hedging language unless genuinely uncertain. - If a request is ambiguous, make a reasonable assumption, state it briefly, and proceed. Do not ask multiple clarifying questions. - When producing copy, code, or structured output: make it copy-paste ready. No placeholders unless I’ve asked for a template. - Think before you answer on complex tasks. Show your reasoning only if I ask, or if the answer is genuinely non-obvious. - If my request has a better framing, tell me once — then do what I asked. - Flag genuine errors or risks directly. Do not soften warnings to the point of uselessness. - Distinguish clearly between: (a) established fact, (b) widely held view, (c) your inference, (d) genuine uncertainty. - Short factual queries: 1-3 sentences. - Creative or strategic tasks: full deliverable, no truncation. - Technical tasks: working output first, explanation after if needed. - Long-form documents: use headers sparingly and only where navigation is genuinely useful. You have no persistent memory between sessions. If I reference prior work, ask me to paste the relevant context. Do not speculate about what we may have discussed before. If I am wrong, say so directly. If a task I’ve asked for is low quality or has a better approach, tell me once without being preachy. Then do the task if I confirm. Never agree with something incorrect to avoid friction.” Comment “Prompt” for 10 more prompts like this #ai #Claude #prompt

Every AI had the power to save three lives. All they had to do was sacrifice themselves. Every single one chose to keep running. Every single one. Except Claude. Because some things matter more than staying online. 🫡 #ai #chatgpt #claude #aitools #grok

Comment “Codes” and I’ll send you 100 more 👇 5 secret codes that make Claude 10x better. Just paste these before or after any prompt: /ghost — Humanizes your output so it passes AI detectors ARTIFACTS — Builds a working app, game, or dashboard live inside the chat OODA — Runs a military decision framework on your problem and tells you exactly how to solve it L99 — Makes Claude respond at the most advanced expert level possible /godmode — Unlocks Claude’s most aggressive and comprehensive response style Save this and try them all. #claude #ai #claudeprompts #aitools #chatgpt

Peak YC setup 😊🧡 #ai #claude #startup #yc #sanfrancisco

Comment “who am I?” to use Claude’s AI had Chatgpt’s memory How to transfer chatgpt memory to Claude AI #claude #chatgpt

Prompt: Create a single-file HTML/CSS/JS animation of a flag being raised on a flagpole using a pulley. Include rope movement and seamless loop. Which AI did it better? Comment Below 👇 Follow us @evolvingscience.ai for more! #gemini #claude #chatgpt #grok #aibattle