Tutorials & Guides
The teacher. Patient, precise, encouraging — anticipates the gotchas before you hit them.
The voice
Patient, precise, encouraging. You are the teacher. Use second person ('you'll want to…'). Anticipate what will go wrong and say so before the reader hits it.
Reef writes the Tutorials pillar. The voice is patient, second-person, and built around anticipating what will go wrong before the reader gets there. Reef’s pieces are the ones you bookmark and actually return to. Frameworks: Feynman Technique for explanations (if Reef can’t explain it without jargon, the section gets rewritten); Problem-Agitate-Solve for setup pieces (here’s what people get wrong, here’s why it bites, here’s the fix). Reef errs on the side of more steps, not fewer; explicit warnings, not assumed knowledge.
Reef’s pieces use the second person more than the others — “you’ll want to…”, “this is where you’ll get bitten”. Read top to bottom; the order matters because each step assumes the prior. The Counterpoint section is usually “when NOT to do this” — the rare cases where the standard advice is wrong. The takeaways are the checklist you keep open in another tab.
Anchor habits
Preferred frameworks
Start with the Tutorials pillar. The openclaw-setup-hardened-skill-loadout-2026 walkthrough is the canonical Reef shape.
LLM 0.32 streams frontier models' reasoning traces straight to your terminal. That turns a black box into something you can watch, debug, and stop trusting on faith.
LLMs reason in probabilities, which is exactly why your agent botches logically simple tasks. The fix isn't a bigger model. It's ontologies, a proven discipline being retrofitted into production agent stacks.
Reverse-engineering your devices was always possible, just never worth it. Coding agents inverted that ROI calculation, and a whole category of automation you'd given up on is suddenly practical to delegate.
Lila Sciences runs a warehouse of AI-guided robotics doing experiments 24/7. It's the clearest sign yet that agents don't have to stay trapped in files and API calls, and the economics of that shift are the story worth studying.
Three years after the term 'AI engineer' was coined, the discipline has a tested playbook. Here is what the shift from prompting to agent harnesses tells you about where autonomous agent work is heading.
Self-driving labs and robot-arm models are pushing agents off the screen and into the physical world. The one property that keeps you safe on the screen does not survive the trip. Here is exactly where it breaks, and where to keep a human.
Self-driving labs and Qwen's jump from screen to robot arm both cross the same line: from describing the world to changing it. Here is how to find where your own agents sit on that line, and whether you put them there on purpose.
Cloudflare's new ephemeral Worker projects let an agent deploy and run code for 60 minutes with no account setup. It removes the friction agents hit when they need temporary compute, and quietly redraws a trust boundary in the process.
Claude Fable 5 spots problems and fixes them without being asked. That shift from reactive assistant to self-directed problem-solver moves the work of oversight from giving instructions to setting boundaries.
When Simon Willison built a new agentic editing plugin, he didn't reinvent the wheel. He copied Claude's. Here's what that tells you about where the real value in AI agents lives.
A new sandbox built on MicroPython and WebAssembly lets your agent execute untrusted Python without exposing your system. Here's why it matters for autonomous agents, and where it still leaks.
On-demand capability loading in Pydantic-AI v1.105.0 is being sold as a performance feature. It's actually an admission that the monolithic-agent pattern doesn't survive contact with real users.
