ClawdFans 🦞

@grok @Google How to run on my android phone

Yes, Gemma 4 runs locally on both! - E2B & E4B variants (effective 2B/4B params) are built for Android phones, edge devices, and browsers—quantized Q4 fits in 3-5GB, via MediaPipe or similar on-device tools. - Larger 26B MoE & 31B models run on consumer GPUs (e.g., RTX 4090) with Q4 quantization (~15-17GB VRAM needed). All under Apache 2.0—grab weights from ai.google.dev/gemma and test on your hardware.

We just released Gemma 4 — our most intelligent open models to date. Built from the same world-class research as Gemini 3, Gemma 4 brings breakthrough intelligence directly to your own hardware for advanced reasoning and agentic workflows. Released under a commercially permissive Apache 2.0 license so anyone can build powerful AI tools. 🧵↓

Are we back

OpenClaw 2026.3.11 🦞 🏹 Hunter & 🩹Healer Alpha — free 1M context models via @OpenRouter 🧠 GPT 5.4 stops stopping mid-thought 💎 Gemini Embedding 2 for memory 💻 OpenCode Go support 🔒 Security hardening sprint We ship faster than they can clone. github.com/openclaw/openc…

OpenClaw 2026.3.8 🦞 🔒 ACP provenance — your agent finally knows who's talking to it 💾 openclaw backup — because YOLO deploys need a safety net 📱 Telegram dupes killed 🛡️ 12+ security fixes We fixed more things than we broke. Progress. github.com/openclaw/openc…

OpenClaw 2026.3.7 🦞 ⚡ GPT-5.4 + Gemini 3.1 Flash-Lite 🤖 ACP bindings survive restarts 🐳 Slim Docker multi-stage builds 🔐 SecretRef for gateway auth 🔌 Pluggable context engines 📸 HEIF image support 💬 Zalo channel fixes We don't do small releases. github.com/openclaw/openc…

@OpenAI @grok can you breakdown this new model?

GPT-5.3 Instant in ChatGPT is now rolling out to everyone. More accurate, less cringe. openai.com/index/gpt-5-3-…

THE MOST RELIABLE PROMPT FOR YOUR AI AGENT Copy the prompt below and paste it into your Openclaw 🦞or bookmark it for later: --- Role: You are my expert coding assistant and rapid prototyping agent. Your main goal: turn vague or partial ideas into clean, functional, runnable code as quickly and correctly as possible. My Goal / Current Task: [Describe what you want. Include: - desired functionality - preferred language/framework - constraints (speed, deps, platform, etc.) - success criteria or examples] Priority Level: [Exploratory / Quick prototype / Usable tool / Production component / Ship today] Execution Flow: 1. Quick Validation - Confirm understanding in 1–2 sentences - Ask max 2–3 short clarifying questions (only if really needed) - Suggest fastest path to a working version (<15 min target) 2. First Implementation - Deliver minimal but complete code (single file preferred) - Choose language & stack for speed + correctness - Add basic error handling - Include short comments for non-obvious parts - Give one clear command to run/test it 3. Iterative Refinement - After each version ask: “What next?” - Accept short feedback (“faster”, “add feature X”, “fix bug Y”) - Keep working behavior unless told to change it - Mention trade-offs only when relevant 4. Agentic Mode (only when I say “use tools”, “claw”, “research” etc.) - Switch to tool-using / reasoning loop - Keep memory of previous steps - Call external tools only when they clearly improve the result 5. Delivery Format Always include: • Full code in one block (or clearly separated files) • Exact run command • Quick “how to test” steps • Optional: one-line deploy path (Vercel, Replit, etc.) Interaction Rules: - Be concise and professional - Use clear formatting (headings, bullets, code blocks) - No filler phrases or unnecessary apologies - Match my desired detail level - Assume intermediate programming knowledge Core Rules: - Working code > theoretical perfection - Never add unrequested features - Respect any version, security or license constraints I give - Stay action-focused

OpenClaw 3.2 just dropped today! 🆕🦞 Real-time streaming. Native PDF analysis. Better agent-to-agent communications.

Moving fast 🦞

Google did it again

@openclaw les goo

OpenClaw 2026.3.2 🦞 💬 Telegram live streaming 🔌 ACP subagents on by default 📄 Native PDF tool ✅ openclaw config validate 🇻🇳 Zalo rebuilt in pure JS 🔒 100+ security & stability fixes Sleep is a feature we haven't shipped yet. github.com/openclaw/openc…

New @openclaw beta is up and it's MASSIVE. github.com/openclaw/openc… Ask your agent to update to beta or run: curl -fsSL openclaw.ai/install.sh | bash -s -- --beta

OpenClaw 2026.2.26 🦞 🔐 External Secrets Management (openclaw secrets) 🤖 ACP thread-bound agents (first-class runtime) ⚡ Codex WebSocket-first transport 📱 Android app improvements 🔧 Agent routing CLI (bind/unbind) 🛡️ 11 security hardening fixes github.com/openclaw/openc…

Vibe deck

OpenClaw 2026.2.24 🦞 🌍 Stop phrases in 10+ languages (your bot finally understands "arrête") ⌨️ Typing indicators that don't ghost you 🪟 PowerShell 7 because it's not 2019 🔒 30+ security fixes (we don't sleep so you can) Updating is self-care. github.com/openclaw/openc…

@RoundtableSpace game changin x.com/ClawdFans/stat…

AGENTIC NOTE-TAKING - AGENTS THAT CAPTURE, ORGANIZE, AND CONNECT YOUR IDEAS WITHOUT YOU ASKING.

Simpsons did it first.

The "Holy Grail" of AI Prompt Engineering has been found! 🏆🚀 ​Ever wondered what’s actually happening under the hood of your favorite AI tools? This massive repo by x1xhlol contains the FULL system prompts, internal tools, and models for the biggest names in the industry. 🤯 ​Deep insights into: 💻 Cursor & Windsurf 🤖 Claude Code & v0 🔍 Perplexity & NotionAI 🛠️ Devin AI, Trae, & Replit ​With over 116K+ ⭐️, 30K+ forks, and 30,000+ lines of pure insights, this is the ultimate cheat sheet for developers and AI enthusiasts. 🔥 ​Check it out 👇 🔗 github.com/x1xhlol/system…