Tag: claude-code
All the articles with the tag "claude-code".
-
Cutting Rules to Save Tokens: One Week of AI Harness Ablation
Most people keep piling instructions onto their AI, but removing rules is what actually saves tokens. Here are the one-week results of my ablation experiment and how Anthropic recommended the exact same thing.
-
My AI's Memory Files Kept Growing, So I Deleted 101 and Added a Gate
After a few months with Claude Code, the memory files it wrote for itself had piled up past four hundred, mostly running logs and client status. Here is how I sorted them, cut them to just over three hundred, and put a write gate in front to stop the regrowth.
- Updated:
Having Claude Dig Through a Month of My AI Chats: Habits, Mods, and Auto Mode Rules
I asked Claude Code to sift through over 300 chat logs from the past month to extract four collaboration habits, turn my most frequent catchphrases into switchable modes, and refine auto-review rules using real rejection cases.
-
Switching Your AI's Working Mode with One Command: Building work-mode and a Week of Testing
I built work-mode, a Claude Code plugin with autonomous, delegate and wrap-up modes plus desktop/phone reply formats. The pain points, how it evolved, numbers from 136 sessions, and prompts you can copy.
-
Measuring Again Three Months Later: From AI Adoption L2.5 to L3
In July I rated myself L2.5 on Boris's AI adoption table. Three months later I measured again: Claude can reach my knowledge base and old chats, code goes through Codex review first, schedules run everything, and 430 sessions in two weeks were started by code.
-
Burning and Saving AI Quota: How I Audited, Allocated, and Spent My Compute
An audit of seven AI quota pools: burning expiring credits on self-verifiable tasks while using surplus compute to build long-term token savings. Combines a terminal dashboard and routing scripts to dispatch tasks dynamically and refine rules from logs. Includes complete tables and reusable prompts.
-
I Asked My AI "Now What?" 41 Times in Two Weeks
Letting an AI run long jobs in the background is easy. Knowing where it got to when you come back is not. I counted two weeks of my own messages and found I was missing a status board, not more monitoring.
-
Which Claude Code Plan to Buy: Pro, Max 5x or Max 20x
Who should buy Pro, who should buy Max 5x, whether Max 20x is worth it, and when to just use the API. Prices come from Anthropic's official pages; the usage judgment comes from running Claude Code every day.
-
Two Claude Mods I Built: a Recording Mask and a Warm-Cache Plugin
I built two Claude Mods: one that masks sensitive info while I record courses, and one that keeps the Claude Code prompt cache warm. Then I found Claude Code already has idle compact built in.
-
How I use herdr: three workspaces for standing, today and long-term
After Opus 5.5 analyzed 30 days of my session logs, I split herdr into standing, today and long-term workspaces. Four status lights keep the standing agent aware, and Opus can jump into the next pane to rescue a task.
- Updated:
My Pre-Trip Remote Checklist: Leaving the Mac Mini as Home Base
Eight things I checked before flying out, to keep a home Mac Mini reachable as the base for Claude and Codex: Tailscale, Herdr, caffeinate, auto-reboot, and remote file access.
-
159 Confirmation Prompts in 14 Days, and About 70% Were False Alarms
I audited 14 days of PreToolUse hook ask prompts: 159 of them, roughly 70% false positives. A guard hook should match the command position, not the whole string, and you can measure the fix from hook events in the transcript without replaying anything.
-
Keys Leak Where You Feel Safe
Three near-leaks of API keys in one week, and none came from pasting one somewhere public: skill packaging, a masking command that failed, and a build config. Plus the rule I use to decide whether to rotate.
-
Running 5+ Claude Code/Codex at Once: My Five Fixes for the Attention Bottleneck
Once I have more than 5 Claude Code/Codex windows going, I hit an attention bottleneck — constant context switching is genuinely exhausting. Five things that get me past it: a good multi-window manager, voice input, an overnight schedule, a reviewable skill, and regular pruning.
- Updated:
When to Call In Fable: Three Moments, and Don't Turn On ultracode
The most expensive model is not meant to run the whole way. Three moments where calling in Fable feels right to me, plus why I cap Opus 5 at med, and the weekly cycle I fell into: efficiency king Monday to Thursday, Fable mode Friday to Sunday.
- Updated:
Your Own Machine Is the Attack Surface: Leftover Credentials, Permission Flags, Untrusted Input
Everyone talks about AI security as whether the model will say the wrong thing. What actually bites you is the plaintext keys left on your disk, the permission flag you waved through, and the untrusted input you fed it.
- Updated:
More Subagents Won't Make You Faster
From wanting 13 subagents to chat for me, to one person running 1000, to agents arguing across a table — the real bottleneck is the human main agent, and the fix is deduping before you dispatch.
- Updated:
No Error Doesn't Mean Success: Silent Failures, Opus Fabricating Tool Output, and the Hook That Finally Stops It
Exit 0, HTTP 200, and a model saying "done" are all unreliable success signals. Cases from my own code, CLIs, and APIs through to a model inventing commit hashes out of thin air, ending with the one thing that actually holds.
- Updated:
7 Practical Tips for Talking to Claude 5: Less Drift, Less Chatter, Less Slacking, Less Nagging
Some practical tips I have been sharing with friends lately on talking to the Claude 5 generation of models: a home-made explain skill for alien-speak, answering everything in one pass instead of letting the conversation drift, voice-input setups, first-principles pruning, a hook that treats slacking, a minimal CLAUDE.md, and a fix for the phantom pushback habit.
- Updated:
Where the Quota Actually Goes: Claude Code Cache Bugs, Opus 4.7's 2x Burn, and Runaway Subagents
Four quota incidents in full: two official cache bugs, an 80x jump in API requests pulled from JSONL, Opus 4.7 measured at 2x consumption, and one session of mine that burned 90% in half an hour.
- Updated:
Two Worlds of Web-Based Claude Code Setup
Web Claude Code has two completely different runtimes — cloud VM and remote-control. Here is what I learned setting up both, the small bugs I hit, and why some user-level configs simply cannot be lifted to the cloud as-is.
- Updated:
A Year In, My Answer to "Terminal or Desktop App?" Has Changed Three Times
From March's Cowork token burn, to April's "still the terminal," to May's "beginners should just start with the desktop app," to August's "the CLI is the firstborn." One question, three different answers across five posts, laid out in the order I gave them.
- Updated:
Fable 5's One-Week Return: Turning the Most Expensive Model into a Skill Distillation Engine
Fable 5 came back for one week only. I did not spend it on daily busywork — I spent it on high-leverage judgment, distilled into skills that cheap models can follow.
- Updated:
The More Rules You Add, the Less Claude Listens: Trimming 36% of Baseline Context with an AI Fleet
Piling on rules makes LLMs miss what actually matters. When a 1M context window gets clogged with noise, you need an overhaul. Here is how an automated team of subagents trimmed my baseline harness by 36% by stripping out narrative fluff and conflicting constraints.
- Updated:
Leveling Up the iPad Workflow — Surviving Internet and Power Outages, and Why I Ditched the Magic Keyboard
A follow-up to the full iPad-runs-Claude-Code guide. This one is not about setup, it is about robustness: home internet in Bangkok drops, the rainy season kills the power, so how does the host machine survive? Plus why I gave up the Magic Keyboard for my mouth and some gestures.
-
Two Skills for Making AI Speak Plain Language: ASD-STE100 for Grammar, ISO 24495 for Structure
A simple asd-ste100 skill I threw together three weeks ago got reshared and picked up three PRs. To make AI speak plain language, pair ASD-STE100 for grammar and vocabulary with ISO 24495 for logic and structure.
-
Information Diet: I Pulled My Own Chrome History and Audited Where My Attention Goes
People obsess over whether every bite of food is clean, then swallow piles of dirty info online without a second thought. So I pulled my own Chrome history db and had it analyze my browsing habits. A few sections came out uglier than I expected.
-
Six New Context Engineering Rules for Claude 5, and the 1,473 Lines I Cut
After reading Anthropic's context engineering guidance for Claude 5, I turned the key points into nine cards, then cut 1,473 lines from a harness I had built up over four model generations.
-
I'm Stuck at Stage 2.5 of AI Adoption
After watching Boris Cherny break down the stages of AI adoption, it hit home: I'm stuck at stage 2.5, where limited attention plus low trust becomes a vicious cycle. This week I used late-night schedules, a morning dashboard, and herdr auto-spawning sessions to cut a 4-5 hour workflow down to 1-2 hours.
-
Two Traps Running Claude Code on Local Ollama: Truncated Context, and A3B Buckling Under Heavy Verification
Two things I logged: cco (Claude Code driven by local Ollama) had long been giving off-topic answers, and the root cause was not a weak model but a full harness whose system prompt had ballooned to 30-50k tokens and was being silently truncated; then I put A3B on the Mac mini for two days as a night worker.
-
Running Taiwan Sports-Lottery Odds With Mobile Claude Code, Standing in the Lottery Shop: the Odds Are Garbage, You'll Lose
Walked past a lottery shop, felt like placing a sports bet, and on a whim wired up a betting-odds skill on mobile Claude Code right there. The verdict: the correct-score market carries a 191% overround, every expected value is negative, you lose.
-
Worried About Agents Going Rogue? Start Isolated, Hand Off Gradually
Someone asked what to do if you worry about AI agents going rogue. My answer: don't pick a highly autonomous lobster up front. Start in an isolated setting where you supervise and approve every step, then let go slowly.
-
Claude Code To-Do Tiering: Auto / Review / Collaborate / Manual
A small discovery from using Claude Code: sort to-dos by repo, then into four tiers — auto, review, collaborate, manual. Stuff that used to drag on all day now finishes in an hour.
-
The Blind Spot in AI Self-Verification — Why You Need an Uninformed Agent to Audit
AI is great at "looking done." To catch its blind spots, you can't rely on it checking its own work — you bring in an auditor that knows nothing, inherits no context, and gets only the rules and the output.
-
Better Context Beats a Stronger Model — repowise's Five-Layer Architecture
I came across repowise, a tool that inserts a layer of context between your codebase and the model. It is a good excuse to talk about a bigger idea: when a model breaks things, the first instinct is to reach for a stronger model, but the real cause is usually too little context.
-
A Student's Seven-Agent Setup That Wouldn't Run — A Cautionary Tale About Harness First
A student built seven agents on an M5 Max and kept hitting timeouts. The problem wasn't a weak model — it was skipping the basics: harness, context management, task routing.
-
Love-Hate With Anthropic — Sneering at the Elitism While Devouring Their Engineering Blog
I have a love-hate thing with Anthropic: I hate the smug elitism and the black-box token-burning, but their engineering blog is some of the best out there, especially the security and permissions stuff.
-
The Claude Code Ecosystem: Skills, Hooks, and Agents in Practice
Claude Code has grown well past a basic terminal CLI into a system powered by skills, hooks, and agent orchestration. Here is how those pieces connect to keep daily workflows dependable.
-
Running Claude Code on iPad — tmux + Tailscale + Moshi Setup Guide
A complete setup for running Claude Code on an iPad 11-inch: tmux for persistent sessions, Tailscale for NAT traversal, and Moshi as the SSH terminal. The goal is rebuilding 90% productivity, not just fixing bugs.
-
Cross-Model Review: Stop Letting AI Grade Its Own Work
When an AI reviews its own output, blind spots are inevitable. Bringing in Codex for an independent pass or running a three-model expert panel with Review Council offers a practical fix.
-
Three moves to firepower-demo Agentic AI at your company—desktop cleanup, conflicting meeting notes, multi-version emails
If you want to make your manager or coworkers feel Agentic AI in one second—and then open the door to an internal training engagement—my three go-to moves are cleaning up messy desktops, reconciling conflicting meeting notes and emails, and producing multi-version drafts for different recipients. These three hit office workers right in the pain.
-
Covey's time management matrix, now with Claude Code—how each quadrant shifts
Covey's time management matrix from The 7 Habits of Highly Effective People splits time on two axes: important × urgent. Drop Claude Code into the picture and each quadrant's time allocation shifts. The long-term compounding from AI isn't in Q1, it's in Q2.
-
Two MCP servers for Taiwan public data—what mcp-taiwan-legal-db and Twinkle Hub are each good for
Two MCP servers wrapping Taiwan's public data sources—one focused on court judgments and statutes, the other aggregating 52,960 government open datasets plus 37 local utility tools. They turn out to be complementary; I now run both.
-
Judging AI Risk on Two Axes: Reversibility × Environment Isolation
A common question in corporate trainings — "can I let AI do this automatically?" I answer with two axes: reversible vs irreversible, isolated vs production. This 2x2 prevents more incidents than any prompt-engineering tutorial.
-
Teaching non-technical people AI: five things the tech world assumes everyone knows
One student was a seasoned HR consultant who managed all her files across several USB drives and had never heard of an API. Here are five knowledge gaps that show up constantly when teaching AI tools outside the tech world — and what actually works.
-
1M context isn't about holding more — it's about managing better
Claude Code's context window grew from 200k to 1M. The instinct is to put more in. That instinct is wrong. Five session management operations, four strategy rules, and the two most common ways to break your own context.
-
Knowledge base deep clean: using AI to find months-old problems
Rules trimmed from 13,020 to 7,590 tokens. 92 broken wikilinks. Five root causes behind a failing todo system. Four days of cleanup, one quarterly SOP.
-
See the Outside, Think the Inside: Two Claude Skills That Pair
I saw a competitor selling well and my first reaction was anxiety. But before letting anxiety win, I wanted to actually understand what it was. Lately I have been pairing two Claude Skills for this — H/V analysis to see the outside, first-principles to see the inside.
-
A Legal Counsel Used Claude Code to Catch Patent Infringement — Days Down to Hours
A legal counsel at a mid-sized company took our Claude Code course, then compressed a multi-day patent infringement audit into a single day — and actually caught a real case that is now in legal proceedings. This story finally turns "Claude Code is not just for engineers" from a slogan into a fact.
-
My Claude Code Automation Stack: /loop, Monitor Tool, the Skill System, Obsidian Reflection Reviews
From /loop clocking in for me at ERP, to Monitor Tool as a background event gatekeeper, to Google Map MCP saving browser clicks, to Obsidian + Mirror framework for weekly reflection. Here's my six-month Claude Code automation stack, plus a clarification of how commands and skills actually split.
-
Six Months with Claude Code: From First Affection to Hesitation, From Overtime to Detox, From Mandarin Circles to the Thai Community
From January to April, my state of mind with Claude Code went from "that PTT-generation affection for the CLI" to "20x hesitation" to "AI detox" to "practicing Thai at a Bangkok meetup." Here's the six-month arc, along with the nights agents worked overtime for me until 5 AM.
-
Two Months of Claude Code Harness Ops: Crashes, Zombies, Cache Misses, and Plan Mode Edges
From March to April, running Claude Code as a long-running harness surfaced a bunch of issues. Survival tactics during the outage, my hook for cleaning up zombie processes, the silent memory bloat fixed in 2.1.86, the resume cache-miss fix in 2.1.90, and the Plan-Mode-question-density problem after 4.7.
-
Practical Claude Code Setup: Slimming CLAUDE.md, Quota Control, and Automated Workflows
When CLAUDE.md swells past a thousand lines, instructions get dropped, and rate limits hit within minutes, your harness needs an overhaul. A practical guide to splitting Rules, Memory, and Hooks, delegating to Haiku subagents, caching rules, Ghostty click fixes, and an automated video pipeline.
-
90% of People Are Two Years Behind: The Real AI Adoption Gap
I demonstrated Claude Code to friends and they were all amazed. Then I gave them free Pro trial passes — three of them. Not a single one was used. That told me a lot about how AI adoption actually works.
-
hermes-CCC: Bringing All 46 Hermes Agent Capabilities into Claude Code
Someone ported the entire NousResearch Hermes Agent into 46 native Claude Code Skills. No OAuth, no external processes, just restart CC and you're ready. The core philosophy is "procedures-as-prompt" — structured markdown instruction scripts drive AI behavior without writing a single line of code.
-
This Week in Claude Code: Four Practical Resources I Bookmarked
From enterprise workflows and local job portals to sports analytics and beginner onboarding, four new Claude Code resources popped up this week. Each highlights a use case that hasn't been tapped enough.
-
Digging into Claude Code's JSONL: Manual Context Usage, Cache Types, and Compact Detection
Claude Code's `/stats` only tells you the total token count, but the JSONL files hide a lot more—cache types, compact triggers, image-triggered cache misses. Here are the useful formulas I collected while building cc-audit.
-
cc-audit: A Skill That Stops Claude Code from Burning Your Tokens
I turned my viral token-saving tips into an automated audit skill for Claude Code. Feed it to Claude Code, and it inspects your settings, recent sessions, rules, and context usage to hand you an actionable checklist of fixes.
-
Make a Claude Code Tutorial Video in 30 Minutes with Remotion Skill
I'd had the Remotion skill installed for a while, mostly just playing with it. Once I started making a free Claude Code tutorial series, I realized how good it actually is—from script discussion to export, 30 minutes gets you a 15-minute video with no retakes.
-
How Claude Code Skills Actually Load: Lazy Loading and a 30-50% Trigger Rate
Skills don't dump all their prompts into context by default. They load layer by layer on demand. But passive trigger rate is only 30-50%, so slash commands are your safest bet.
-
When Your Primary AI Could Fail at Any Moment: Multi-Endpoint Fallbacks and Setup Sync for Non-Coders
When your daily work relies heavily on a single AI provider, downtime or sudden quota cuts can halt operations overnight. Here is how I evaluate fallback models for non-coding tasks, sync rules and memory across tools, and stay ready to switch.
-
The 5 Levels of Claude Code: You Don't Choose to Upgrade, the Ceiling Forces You
From Raw Prompting to Orchestration — the 778-upvote Reddit framework where every level exists because the previous one broke.
-
Turning Books into Claude Skills: From Textbooks to Atomic Habits, the World Is Already On It
Turning books, frameworks, and methodologies into Claude Skills — making knowledge executable as interactive coaches. A roundup of popular examples and practical lessons.
-
The Real Next Level in Claude Code: Context Management and Hooks
From Claude.md to Hooks — sharing the context management upgrade after heavy daily Claude Code usage. When rules pile up and compliance drops, Hooks become the perfect gatekeeper.
-
Obsidian + Claude Code: My Daily Knowledge Management Workflow
Put your entire life into Obsidian, then let Claude Code tap in. How I do daily logs, manage to-dos, and why local Markdown beats Notion.
-
Claude Code Workshop Prep Notes: Teaching Non-Coders to Go Terminal-First
Designing a Claude Code workshop for non-engineers — from analyzing signup demand to crafting the curriculum, plus 15 quick-start tips for zero-coding-background learners.
-
claude-work-timer: Auto-Track Your Claude Code Working Hours
Billing clients by the hour with Claude Code? I built an open-source plugin to track actual working time automatically.
-
Claude Code Initial Prompts: Six Practical Habits for Working with AI
Six practical habits for writing Claude Code initial prompts: clarifying goals, mapping manual workflows, picking the right model, and leaning on plan mode.
-
What I Learned from 7,843 Claude Code Conversations
46 days, 6 projects, 7,843 conversations. This is not a manual — it's a diary of trial and error.
-
I scanned 500 Claude Code sessions and the AI stopped making the same mistakes
I used claude-log to scan 500+ Claude Code sessions across 6 projects, extracted every time I yelled at the AI, and turned those into CLAUDE.md rules. Here's what I found.
-
FIRE Planning with Claude Skills: Talking Retirement with AI
I fed all my assets, liabilities, income, and expenses into Claude Code, then just asked "am I going broke?" A month of conversations turned into a full FIRE planning toolkit — and why I open-sourced it.
-
The Claude Code Dunning-Kruger Curve: From Complete Stranger to Total Sync
A non-engineer power user's journey through the Claude Code Dunning-Kruger curve: from day-one confusion and day-three hubris to week-two wall-hitting, before mastering rules, skills, and multi-model collaboration.