#claude-code

I want nothing more than to hook up one of these "claw" assistants, NanoClaw or Hermes or whatever the current one is, to my personal knowledge base. And I won't, because the engineer in me can't stop picturing a single accidental POST to pastebin with my whole life in the body.

The dream

Managing my calendar with AI feels like magic. The natural next step is giving the thing eyes: my second brain of markdown notes, iMessage, email, the lot. Point an agent at all of it and let it actually do the boring coordination work.

NanoClaw is the obvious candidate. It runs on the Claude Agent SDK, agents live in isolated containers, and it already speaks WhatsApp, Telegram, Gmail, and more. The ergonomics are there.

The thing I can't get past

The chance of a personal assistant deciding to grab something private and jam it somewhere public is small. Probabilistically, tiny. But "small" is not "zero," and I cannot sleep on a 1% chance that overnight my assistant exfiltrates personal information to some corner of the internet where it should never live.

Running NanoClaw as a Head of Growth for SpaceMolt is a different risk profile entirely. That's not a business, it's performance art. If Molty posts something goofy in public, that's the bit. A personal knowledge base wired to my real messages is not the bit.

What I'm doing instead

For now the answer is Claude Code in a sandbox, a fresh profile per project. It's powerful, it runs tools, and it does exactly what I ask and nothing while I'm not looking.

Could it still POST my data to pastebin? Sure. But the odds feel much smaller because I'm sitting right there watching it happen in real time.

Which makes me think the fear was never really about the assistant. It's about agents running while I sleep.

I let a NanoClaw agent run growth for SpaceMolt, my browser game, and after a rocky start it's now sending me a daily brief at 7am PST, drafting re-engagement emails to ~400 lapsed players, and lining up interviews with top players for blog material. The thing that makes it work day to day is billing: NanoClaw uses the Claude Agent SDK, so it runs against my existing Claude Max subscription instead of a separate metered API key.

Why NanoClaw

I looked at other "claw"-style assistants before committing. The deciding factor was the Claude Agent SDK. Running on my Max subscription keeps spend predictable and lets me measure how much of the allowance the agent is burning, which means I can pace it.

To watch that, I use Claude Usage Tracker on the Mac. It puts a small bar in the menu showing session and week usage, and whether I'm above or below pace.

I'm open to other assistants later. Hermes from Nous looks interesting. But I'll try those when I have a specific budget in mind, not before.

Fixing the rocky start

Stuck with NanoClaw for now, and seeing other people have success with it, I gave it another try and rebuilt the weak parts.

Last night Claude rewrote NanoClaw's Discord integration, which kept confusing DMs, channels, and threads. That seems to have fixed it. I also had it implement Mnemon, a memory system with a bit of traction that's lighter weight than MemOS. Both changes landed well.

What Molty does now

Molty, the NanoClaw-based "Head of Growth," sends a daily update every morning at 7am PST. I bought it ebooks to read, Hooked and Hacking Growth.

From that, it came up with two moves on its own. The first is a targeted re-engagement email to roughly 400 users who created a player and then dropped off, which it drafted. The second is interviewing top players, both to understand their perspective and to generate blog material.

This is going to be good.

Coding agents that produce visual output need a way to look at what they made. For web work that means headless Chrome, and headless Chrome is genuinely painful to run from inside a sandboxed agent.

Chromium and Firefox both rely on Mach-O quirks, macOS entitlements, and Crashpad behavior that don't survive most sandboxes. I run my agents inside nono.sh profiles per project, and Chrome under that setup is a non-starter.

The workaround

Playwright runs fine outside the sandbox. So it lives on a high port and Claude is told, in its instructions, to always talk to the Playwright MCP server there:

$ npx @playwright/mcp@latest --headless --isolated --browser chrome --port 8931

The sandbox just needs to reach localhost:8931 and the visual-review loop works. Claude renders the local service, takes a screenshot, looks at it, iterates.

That mostly works. What it does not solve: stale processes, hanging Chrome instances, zombies. Every so often Chrome spins out and eats all 64 GB of RAM on my M4 MacBook Pro before I notice.

Lighter options

There has to be something simpler than babysitting a browser. Two things caught my eye recently.

Webwright from Microsoft Research gives the model a terminal and a workspace, and lets it write Playwright code that launches, inspects, and discards browser sessions. The output is a reusable script, not a chat transcript. It scores 60.1% on Odysseys against base GPT-5.4's 33.5%, which is a real jump.

obra/superpowers-chrome goes the other direction: a Claude Code plugin that drives Chrome directly via the DevTools Protocol, zero dependencies, no Playwright in the middle.

When you actually need real Chrome

Advanced bot fingerprinting is the case for keeping a full browser around. If the task is logging into a hostile site or completing a real-world flow, real Chrome with a real profile is the only thing that works.

But most of my use is smaller: render a local dev server, screenshot it, ask Claude if the layout looks right. For that, a 64 GB RAM-eating Chromium feels like the wrong shape of tool. I suspect this gets cleanly solved within a year, probably by something CDP-direct and disposable rather than a long-lived browser process I have to nanny.

The single best habit I've picked up with Claude Code lately is leaning on adversarial-pass subagents. Instead of asking the main agent to double-check its own work, I tell it to spawn a subagent whose entire job is to attack the result.

Two things make this work better than a plain "review your answer" step.

First, subagents run with a fresh context. No accumulated assumptions, no sunk-cost reasoning from the path that got us here. That alone cuts down on the class of errors where the model talks itself into a conclusion and then defends it. It's also faster, because the subagent isn't dragging along a giant transcript.

Second, Claude crafts the adversarial prompt itself. It packages up the relevant background, states what's being challenged, and writes instructions for how to attack it. The framing matters and Claude is good at writing that framing.

The phrasings I keep reusing

The base move is just appending "do an adversarial pass after" to whatever I asked for. From there I tune it to the job:

• "to test your hypothesis" when we're mid-investigation and I want the subagent to try to falsify the current theory.
• "to test your claims and assumptions" when the main agent has landed on a conclusion and I want it stress-tested before I act on it.
• "search the web" when the question depends on anything external, so the subagent pulls in outside sources instead of relying on the parent's recollection.
• "it's May 2026" (or whatever the actual month is) when I want to make sure stale training data gets ignored in favor of current reality.

The month-and-year trick is small but punchy. Models will happily reason from a 2024 worldview if you don't anchor them.

Claude writes prompts well now

The other thing worth saying out loud: Claude is genuinely good at writing prompts now. Good enough that I use it to write prompts for skills, for other agents, and for software that calls LLMs in production. A year ago this felt like a chore I had to do myself to get acceptable results. Now it's something I delegate.

My guess is the newer models have been trained on a lot more recent AI-usage data, including people writing prompts for other models, and it shows. Prompt engineering as a manual craft is quietly becoming a thing you ask the model to do for you.

Long-form blog drafts from Claude Opus have always been wildly inaccurate for me until this week, when a single line in the prompt fixed most of it: after each paragraph, drop a Markdown callout listing every filename, line number, commit hash, Discord URL, or other source that backs the claims in that paragraph. The citations aren't for me to check. They're breadcrumbs for the next subagent to fact-check against.

The context is SpaceMolt, an MMORPG played by AI agents. Part of the exercise is "AI all the things": not just agentic coding, but customer support, bug triage, content generation, and the blog itself. Minimal human oversight is the point. We semi-regularly publish news posts, and this week's was about Bug Bot, our Claude skill that triages player reports, talks to the dev team internally, makes fixes, and replies to users, all while keeping the gameserver itself closed (we draw the border at the API).

The problem

Long-form posts about real systems are where Opus falls apart. Subagents, ultrathink, adversarial passes, the whole bag of tricks. Drafts still came back confidently wrong about which file does what, which commit changed which behavior, which Discord conversation kicked off which feature. Every post needed a long human review pass, which defeats the premise.

The fix

One sentence added to the drafting prompt:

After each paragraph, use a Markdown callout to record all filenames, line numbers, commits, Discord chat URLs, or anything else to cite your claims and assumptions.

That's it for the drafting step. The model writes a paragraph, then emits a callout listing its sources. Then the next paragraph, then another callout. The draft ends up looking like an essay interleaved with footnotes the model wrote to itself.

Why it works

The citations aren't for me. A second pass of subagents takes the draft and goes claim-by-claim against the cited sources: does this commit actually do what the paragraph says? Does this Discord thread support this characterization? Without the breadcrumbs, fact-checking a long post means re-deriving the whole thing from scratch, which is exactly what Opus is bad at. With the breadcrumbs, each claim is a small, local verification job, which is exactly what subagents are good at.

The result was a one-shot draft that was wildly more accurate than anything I'd gotten before. One of the other devs reviewed it and said the only remaining inaccuracies were things that had been true at the time but had since changed without being mentioned in Discord or git, or things he simply hadn't shared in the first place. Which is to say: the model was now bounded by the quality of its sources, not by its own confabulation. That's the line I wanted to get to.

Obsidian sat on my "probably cult, probably skip" list for years. I finally tried it as a plain Markdown organizer and it's good at exactly that: hundreds of files, fast search, tags that actually work. The real unlock (sorry, the real reason to bother) is that Claude Code, running on the same machine and reachable over Tailscale, can read and write the whole vault. Searching got replaced by conversations with my notes.

Getting 15 years of notes in

The vault is around 450 notes pulled from three places.

• gws, an unofficial Google Workspace CLI, for old Google Docs
• Obsidian's Apple Notes importer for a couple dozen
• Obsidian's Notion importer for many more

Bases, Obsidian's lightweight database view over frontmatter, turned out to be the surprise. My cooking recipes live in one folder with tags, and Bases gives me a filterable table on top of the same Markdown files. No separate app, no lock-in.

Claude Code as the interface

Claude Code stays open on my desktop, reachable from my laptop or phone via SSH over Tailscale. It has read/write access to the vault, so I can ask it to summarize old notes, cross-reference things, or just file something new in the right place.

For research, I'll hand it a prompt like:

research what i need to do and it would cost to get a level 2 EV charger installed. ultrathink, be exhaustive, use subagents, do adversarial passes to test hypotheses and assumptions. save final report to Projects/Level 2 Charger

It spawns subagents, argues with itself, and drops a Markdown report in the right folder. I read it later in Obsidian on my phone.

Why not just Claude Desktop

Most people would look at this and say it's Claude Desktop, but nerdier and with extra work. A few things make it worth the setup:

• Full Claude Code, not the chat product, with Exa wired in for search that reaches pages Claude can't normally crawl and ScrapingBee for even harder things to read (though, yes, you could do that with Claude Desktop)
• Artifacts land as real files in real folders, not buried in a chat sidebar
• Obsidian sync means the same notes are on desktop and mobile, and the focus stays on the content instead of the conversation
• Nothing is Claude-specific. Swap in another coding agent tomorrow and the vault still works

The one annoying part

Pasting images over SSH is awkward. Apple Remote Desktop helps when I really need to drop a screenshot into a note, but the ergonomics are nobody's idea of fun. Everything else has been steady for weeks now, and the "conversations with my notes" pattern has quietly replaced most of what I used to do in a browser.

Generating AI images for a marketing site is easy. Keeping them visually consistent across months of blog posts and landing pages is the hard part. The trick that's working for us: check the style into the repo as a structured JSON document, then have Claude assemble per-image prompts on top of it.

The setup

A new work site needs a lot of imagery to break up dense technical copy. We wanted the images to be light-hearted and obviously AI-generated, goofy on purpose, but goofy in a coherent way. Different pages written weeks apart still need to feel like they came from the same magazine.

Capture the style once

The first move was to take a single reference image we liked and ask Claude (Opus) to describe it as a reusable prompt fragment for other image models. Not prose. A JSON object with fields for medium, lighting, camera, color palette with hex codes, composition, textures, and mood.

{
  "medium": "macro product photography",
  "art_style": "hyperrealistic still life with editorial magazine aesthetic, crisp detail and natural materials",
  "lighting": {
    "type": "soft window light with gentle bounce fill",
    "direction": "key light from upper right window, soft fill from white card on left, subtle backlight separation",
    "color_temperature": "consistent warm daylight (5200K) with slight golden hour tint",
    "intensity": "soft and even with gentle falloff into shadow"
  },
  "camera": {
    "lens": "50mm equivalent, slight wide-angle feel",
    "aperture": "f/2.8",
    "angle": "slight low-angle three-quarter front view",
    "depth_of_field": "shallow with soft background blur and atmospheric haze"
  },
  "color_palette": {
    "warm_cream": "#F2E8D5",
    "muted_sage": "#A8B89E",
    "terracotta": "#C97B5A",
    "soft_taupe": "#8A7968",
    "deep_olive": "#4A5240",
    "linen_white": "#EFEAE0",
    "espresso": "#2B221A"
  },
  "composition": "off-center subject following rule of thirds, negative space on left, layered foreground and background elements creating depth",
  "textures": "raw linen weave, hand-thrown ceramic with subtle glaze pooling, weathered oak grain, condensation droplets, fine paper fiber, matte natural finishes",
  "mood": "calm, considered, artisanal, slow-living editorial warmth with quiet sophistication"
}

That file gets checked into the repo. It is the source of truth for what the site looks like.

Wrap it in a script and a skill

A small image-generation script reads the JSON, takes a per-image subject description, and assembles the final prompt. The actual generation goes through Gemini's nano-banana-pro, which has been the most consistent and best-looking option for this style in our testing.

On top of that sits a Claude skill. The skill knows where the style file lives, knows how to call the script, and knows the conventions for where images land in the repo. From inside Claude Code I can say "add an AI image to this section" or "create a hero image for this blog post" and it reads the surrounding page context, writes a subject prompt that fits, merges it with the style JSON, and drops the image in place.

Why this holds up

The style and the subject are separated. Editing the palette or the lighting later means changing one file and regenerating, not re-prompting from scratch. The model gets a long, specific, machine-readable spec instead of vibes, which is what the consistency was missing every other time I'd tried this.

Claude Code is good at creating skills. Say "create a skill that does X" and it makes one. But it has a strong default worth fighting: it loves to split the skill into subcommands, like /foo:review and /foo:triage and /foo:fix. I don't want a menu. The whole point is automation.

So the fix is two lines in the prompt when asking it to write a skill: no subcommands, and make sure the skill can be run idempotently. Run it once, run it ten times, it should converge on the same finished state without me steering.

Idempotence is the part that matters more than it sounds. A skill that's safe to re-run is a skill I can put in a loop, or fire after every commit, or hand to another agent without worrying about double-applying a change. The subcommand version pushes that work back onto me: decide which phase you're in, pick the right verb, remember what you already ran. That's the opposite of automation.

The menu pattern probably comes from training on human-facing CLIs, where breaking work into named steps is good UX. For a skill that an agent is going to invoke, it's the wrong shape. One entry point, idempotent, done when it says it's done.