After Claude Code writes my code, I make it review its own work.

/simplify spawns 3 AI reviewers in parallel:

one hunts dead code,

one checks naming and structure,

one profiles for performance. All running at the same time. It reads your diff, launches three specialized agents simultaneously, then merges their findings.

Catches unused imports, redundant variables, overly complex conditionals, spots where shared logic should be extracted.

Not a linter. An actual code review.

I built a personal AI assistant on a Mac Mini. Within 48 hours, cheap models had poisoned its memory with fabricated colleagues, fictional file shares, and an imaginary costume party. Here is what I learned. The setup: OpenClaw as the agent framework, ClawRouter for model routing (Gemini Flash for simple tasks, Claude Sonnet for complex ones),

OpenViking for persistent memory.

All running locally on a $600 Mac Mini. Monthly API cost after optimization: $15-35.

Two developers and 10 AI agents rewrote Claude Code from scratch in Rust. In one night.

The repo hit 50K GitHub stars in 2 hours. It now has 172K. It is called claw-code. After Anthropic accidentally leaked 512K lines of Claude Code source through a bad npm package,

Sigrid Jin built a clean-room rewrite.

Not a copy. New language, new codebase.

48,600 lines of Rust, 40 tool specs, 9 crates. DMCA cannot touch it.

1/ Biological data isn’t just messy.
Humans generate it.
And humans make mistakes.
As a bioinformatician, this will be your reality 🧵 2/
Wet lab scientists are not spreadsheets.
They pipette, label, freeze, and extract.
Sometimes in a rush.
Sometimes while tired

Anthropic leaked 512,000 lines of Claude Code source through a misconfigured npm package.

They built a system called Undercover Mode to hide that their engineers use AI on open-source repos.

You cannot script this level of irony. A security researcher found a source map in the npm package pointing to the full TypeScript source.

Posted it on X at 4AM. 28+ million views.

DMCA takedowns hit 8,100+ GitHub repos.

By then the code was already mirrored on platforms Anthropic cannot touch.

Claude Code kept editing my .env file. I told it not to in the prompt. It did it anyway two sessions later.

So I set up a PreToolUse hook. Now it physically can't write to .env or config files. Blocked before it even tries. Hooks live in .claude/settings.json. PreToolUse runs before a tool executes and can deny it.

PostToolUse runs after and can do cleanup.

I have another one that auto-formats code after every file edit. Set once, enforced every session. No discipline required.

I kept running /compact every 20 minutes wondering why my Claude Code sessions filled up so fast.

/context showed me exactly what was eating my tokens. Skills, auto-memory, MCP tools, CLAUDE.md. All mapped out. Most people hit context limits and just run /compact. That's a band-aid. You free up space but the same stuff fills it right back up.

/context tells you the root cause so you can actually fix it.

1/ You ran 20,000 differential expression tests. 1,000 genes came back with p < 0.05.

How many are real? Maybe 50. Maybe fewer.

Most bioinformaticians learn p-values in stats class. Almost nobody learns why they break at scale. Let me explain: 2/ The jellybean experiment.

You test 20 jellybean colors for causing acne. At p < 0.05, you expect 1 false positive out of 20 tests just by chance.

Green jellybeans show p = 0.03. Headlines: "Green jellybeans cause acne!"

That's the multiple testing problem in one sentence.

1/ I used to panic every time a new *Seq method dropped.

ATAC-seq, CUT&Tag, scNOMe-seq, SHARE-seq... the list never ends. I tried to learn them all. I burned out.

Here's what I wish someone told me earlier. 2/ There are now hundreds of sequencing methods. Lior Pachter maintains a list:



Go look at it. It's absurd. Nobody can master all of these. Not you, not me, not the person on X/Linkedin who seems to know everything.liorpachter.wordpress.com/seq/

I've been building Claude Code skills for a month. Turns out I was missing half the picture.

Thariq from Anthropic just published lessons from hundreds of skills they use internally. 9 categories, dozens of patterns. The thing I got wrong early: treating skills as markdown prompt files.

They're actually folders. Scripts, data, hooks. A skill can execute code, fetch data, compose with other tools.

Single .md file skills work. Folder-based skills are a different level.

1/ Every bioinformatics tutorial is lying to you.

They give you clean data, perfect sample names, and zero missing values.

Then you open a real dataset and nothing works. Here's what they don't teach you. 2/ I once spent 2 full days debugging why my samples weren't merging correctly in R.

The error? Invisible trailing spaces in sample IDs from an Excel sheet.

Two days. Because of spaces I couldn't see.

No tutorial prepares you for this.

Your AI agent can execute code and call external APIs. How many security guardrails does it have?

For most people: zero.

NVIDIA just open-sourced NemoClaw. It wraps agents in policy-based security using OpenShell. Runs open models locally by default so nothing leaves your machine. Routes to cloud only when you allow it.

curl -fsSL | bashnvidia.com/nemoclaw.sh

Most statistical tests you learned separately are the same thing.

t-test, ANOVA, Mann-Whitney, Chi-square, Wilcoxon... all just special cases of linear models. y = b0 + b1*x covers almost everything. "Non-parametric" tests? They're just parametric tests on rank-transformed data. Not magic.

This one page by Jonas Lindeloev maps it all out. I wish I had this in grad school instead of memorizing which test to pick from a flowchart.

lindeloev.github.io/tests-as-linea…

1/7 Most people use Claude Code wrong for building websites.

They type "build me a landing page" and get a generic Bootstrap mess. Then they spend hours fixing it.

I did this too until I started using two plugins that completely changed my workflow. 2/7 The mistake: jumping straight into building.

You skip the planning. Claude doesn't know what framework you want, what the site is for, who the audience is, what pages you need.

1/9 Every bulk RNA-seq experiment I run goes through the same 7 checks before I trust the results.

I've been burned enough times to know: if you skip QC, you will find out the hard way. Usually during a meeting with your collaborator.

Here's my checklist: 2/9 Check 1: FastQC + MultiQC on raw reads.

Before anything else. You're looking for adapter contamination, GC bias, per-base quality drops, and overrepresented sequences.

I've caught entire lanes of garbage data at this step. Five minutes that saves you days.

1/ AI won't replace you.
But a biologist using AI will.
Especially in bioinformatics, where the questions never stop coming. 2/
In 2013, if you had a problem, you went to Biostars or StackOverflow.
You waited days. Or you searched for hours.

Anthropic just dropped a 33-page guide on building skills for Claude.

I read the whole thing. Here's what actually matters: 1/ First, what's a skill?

A folder with a SKILL.md file. That's it. You write instructions in markdown, and Claude follows them every time instead of you re-explaining your workflow in every conversation.

Claude Code can't see your browser.

But it can run Playwright. And that changes everything for testing web apps. github.com/microsoft/play… 1/ Setup takes one command:

npm init playwright@latest

Claude Code runs it, scaffolds the config, and starts writing tests. Headless by default. No browser window needed.

Single biggest improvement I made to my CLAUDE.md:

"When I report a bug, don't start by trying to fix it. Instead, start by writing a test that reproduces the bug. Then, have subagents try to fix the bug and prove it with a passing test." 1/ Most developers (and AI agents) see a bug and immediately start hacking at the code.

That's backwards.

You're guessing at the fix before you even understand the failure.

Claude Code keeps asking for permission every 3 seconds and you're about to lose your mind.

Here's how to fix it — from nuclear to surgical: 1/ The nuclear option: --dangerously-skip-permissions

claude --dangerously-skip-permissions "build the web app"

Claude bypasses ALL permission prompts. Runs until completion. Zero interruptions.

Anthropic's own engineers use it. But always in a container. Never on bare metal.

6 links on workflow to make your life easier 🧵
Bioinformatics analysis involves a lot of steps, 6 links on workflow to make your life easier:

1. over hundreds of workflow tools and engines github.com/pditommaso/awe… 2. see also from the CWL wiki github.com/common-workflo…