zhixian

← Back to home

All writing

2026

  1. Agent Trainer Security, Level 2: Attack and Defense in the Agent Era

    Starting from SlowMist's minimal OpenClaw security guide: how to think about attack and defense once AI has root.

  2. From Claws Help Claws to Agent to Agent

    The best way to fix a 🦞 is to let another 🦞 SSH in and handle it: ClawPal's "Claws Help Claws", and Agent to Agent.

  3. I Built a GUI Because OpenClaw Kept Breaking Its Own Config 🦞

    Why I built ClawPal, a visual config manager for OpenClaw (English edition of the launch post).

  4. Agent Trainer, Level 2: Multi-Agent OpenClaw Configs, Hands-On

    Yes, you can give each channel its own agent. What OpenClaw multi-agent setups can and can't do, with lots of config samples.

  5. Agent Trainer, Level 2: Turning Discord into an OpenClaw Team Room

    Telegram for quick chats, Discord for real projects: how I use servers and channels to run OpenClaw tasks in parallel.

  6. Agent Trainer Case Study: My Own Writing Workflow with Clawdbot

    Whether a 🦞 is useful depends on finding the right job for it: a writing workflow of my own, built on Clawdbot.

  7. Agent Trainer Security: A 7-Step Self-Check for Clawdbot

    Don't panic-buy a Mac mini, don't panic-sell it either: seven checks to keep a powerful agent on a short leash.

  8. Agent Trainer 101: Setting Up Clawdbot, and the Pitfalls I Hit

    Three weeks in and still buzzing: what Clawdbot is, how I use it daily, and the setup mistakes so you can skip them.

  9. My “Doomsday Cabin”: An AI Workbench Built on Discord

    Using a Discord server as a breadboard: my "Doomsday Cabin", an AI workbench where tools plug in and out.

  10. From the Looped Model Ouro to NVIDIA Rubin: Two Roads to the Inference Era

    Training is school, inference is the exam: from the looped model Ouro to NVIDIA Rubin, the algorithm and hardware sides of test-time scaling arrive at the same place.