15:I[981,["658","static/chunks/658-a698693f56d0116a.js","183","static/chunks/183-6beeeb801c274aaa.js","474","static/chunks/app/%5Btype%5D/layout-631c4730b83c8223.js"],"Posts"] 16:T71b,I spent a few days poking at Clawbot and it clarified something I’d been hand-waving: the hardest part of “useful autonomous agents” is increasingly not model quality, but the runtime that keeps the whole thing coherent when tools/LLM calls are slow, flaky, or non-deterministic.

Clawbot reads less like “an agent” and more like a platform:

• A gateway/daemon that owns processes and isolates the system from expensive/unstable LLM calls

• Event-driven orchestration (webhooks) so agents react to real state changes rather than polling

• Fine-grained permissions during onboarding + handshake flows that make “local authority” safer to reuse

• Local-first execution so you can reuse existing tools (terminal, local apps, Claude Code, etc.) without turning everything into a cloud integration problem

This also explains why hackathon agent demos (including one I built last year) often feel brittle: they’re usually an “agent loop + tools,” but they lack the boring runtime pieces—resource management, concurrency control, failure handling, state persistence, and permission boundaries—that let tasks run for long periods without weird interference.

A few questions I’m left with:

1. What’s the right minimal “kernel” interface for agent runtimes (state, scheduling, authority, audit)?

2. How should these systems implement preemption / cancellation / checkpointing for long-running tool chains?

3. What’s the best pattern for “LLM as an unreliable subprocess” (retries, circuit breakers, sandboxing, determinism contracts)?

4. For multi-node setups, what’s the cleanest model: one daemon per node + one gateway, or fully federated gateways?

Full thoughts:

https://chenyu-li.info/blog/15/en/

Where did southern Australia's record-breaking heatwave come from?

Tesla's unsupervised robotaxis vanish after earnings announcement

Show HN: EU Regulations MCP Server – Query GDPR, Dora, NIS2 from Claude

Notes on Clawbot: agents are starting to look like OS runtimes design now

Federal court rules Norfolk's license plate reader system is constitutional

Energy based AI reasoning model – Sudoku solver performance comparison

Show HN: BoxLight- a macOS Sonoma ring light app that doesn't waste space

Choosing a web host feels a lot like a first date

We Optimized the World to Death

The Three-Cent Problem

China's Four-Year Energy Spree Has Eclipsed US Power Grid

Show HN: MoaV – Mother of all VPNs

Stoicism-as-a-Service (SaaS)

Add Support for PyCapsule to Pyspark

Selectively Disabling HTTP/1.0 and HTTP/1.1

We instrumented GitHub Actions. Here's what GitHub won't show you

Predictive Translation: High-Perf Buffer Management Without the Tradeoffs [pdf]

Dark Triad: Explained

I'm back to building my own digital music collection

The Jevons Paradox of AI and New Industries of Creativity

Native Linux VST plugin directory

New Books Aren't Worth Reading

Don't Be a Tourist in Your Own Codebase

"A bootstrap chain for NixOS, builds the system from hand-auditable binary seed"

GOG now using AI generated images on their store

Feds maintain 3.5 – 3.75% rate

Bertrand Russell on Apricots

Conserved shifts in sperm small non-coding RNA during mouse and human aging

AI's HAL 9000 Problem, and What It Portends for the Future

Native Instruments enters into insolvency proceedings

Show HN: Deploying Helm charts without fear