Blog

Articles, experience reports and news about Claude Code.

Claude Code 2.1.225 and 2.1.226: headless OAuth token bug fixed, claude agents now asks for trust

2.1.225 fixes a bug that broke headless sessions by silently swapping your long-lived OAuth token for a short-lived one, adds a workspace trust prompt to claude agents on untrusted directories, and patches several rough edges in the brand-new cross-session messaging. 2.1.226 is maintenance only. Here's what matters in these August 8 releases.

claude-codechangelogoauthheadlessself-hosted-runner

Claude Code 2.1.224: sessions can now talk to each other, a real sandbox bypass closed

2.1.224 fixes a sandbox rule that could be silently bypassed on Linux and macOS, adds cross-session messaging (SendMessage, ListAgents), and ships self-hosted runners for Team and Enterprise. Here's what matters in this August 7 release.

claude-codechangelogsecuritysandbox

Claude Code 2.1.223: four permission bypasses closed, /review merges into /code-review

2.1.223 fixes four distinct ways to sneak commands past permission checks (Bash, invisible Unicode, workflow sandbox, bypassPermissions), folds /review into /code-review, and tightens 1M context window enforcement. Here's what matters in this August 6 release.

claude-codechangelogsecuritypermissions

Claude Code 2.1.222: worktree isolation actually holds now, ultraplan is gone

2.1.222 closes a real isolation gap between worktree sessions and the main checkout, fixes a way background tasks could bypass tool restrictions, and removes the ultraplan feature. Here's what matters in this August 4 release.

claude-codechangelogsecurityworktreesessions

Claude Code 2.1.221: the Bash zsh permission bypass gets closed, background sessions start pushing on their own

2.1.221 fixes a real permission bypass on Bash and PowerShell, adds a credential-masking mode for the Linux sandbox, and changes the default behavior of background sessions: they now commit and push on their own. Here's what matters in this release.

claude-codechangelogsecuritysandboxsessions

Claude Code 2.1.219 and 2.1.220: Opus 5 becomes the default, subagent nesting makes a comeback

2.1.219 makes Opus 5 the new default Opus model, brings back subagent nesting up to three levels deep just three days after turning it off, and adds a real network lock for sandboxed commands. 2.1.220 is a maintenance release only. Here's what matters.

claude-codechangelogopus-5subagentssandbox

Claude Code 2.1.218: /code-review moves to the background, /deep-research stops launching itself

2.1.218 changes the default behavior of /code-review (now a background subagent) and /deep-research (manual trigger only), closes a hole on untrusted agent hooks, and fixes a left-arrow shortcut that could discard a conversation with no way back. What actually matters in this release.

claude-codechangelogcode-reviewsecurityagents

Claude Code 2.1.217: Subagents No Longer Chain Automatically, and a Workspace Escape Is Patched

2.1.217 caps concurrent subagents at 20 and turns off nesting by default, fixes a symlink-based workspace escape and a startup hang risk from brace-heavy frontmatter. Here's what actually matters in this release.

claude-codechangelogsubagentssecuritysandbox

Claude Code 2.1.216: Symlink Security Holes Patched, and Long Sessions Finally Stop Crawling

2.1.216 closes several symlink and git-isolation security holes, fixes a quadratic slowdown in long sessions, and restores background agents' identity on resume. Here is what actually matters in this release.

claude-codechangelogsecurityworktreeperformance

Claude Code 2.1.215: /verify and /code-review No Longer Run on Their Own

Claude Code 2.1.215 removes the automatic triggering of the /verify and /code-review skills after a code change. You now have to type the command yourself. Here's what that means for your review workflow.

claude-codechangelog

Claude Code 2.1.214: The Great Permission Lockdown

2.1.214 closes about a dozen permission holes (PowerShell, Bash, Docker, file commands) and changes what your dir/** rules actually match. It also adds EndConversation. Here is what matters for your setup.

claude-codechangelogsecuritypermissions

Claude Code 2.1.212: /fork Now Goes Background, and Runaway Loops Finally Have a Ceiling

2.1.212 redefines /fork, caps subagents and web searches per session, and auto-backgrounds slow MCP calls. It also closes a plan mode permission hole. Here is what actually matters for your workflow.

claude-codechangelogsubagentsmcpplan-mode

Everyone Teaches You to Use AI. Nobody Tells You How to Make Money With It

ChatGPT, Claude, 20 to 200 dollars a month. The real question isn't how to use AI, it's how to make your subscription pay for itself. A tier-by-tier guide, backed by real numbers.

ClaudeClaude CodeROIfreelanceAIsubscription

Saving Tokens on Claude Code: Stop Blowing Your Quota

Burning through your Claude limit in two days? It's not a tier problem, it's a configuration problem. 7 settings to recover 50 to 60% of your quota.

Claude Codetokensoptimizationcostsproductivitybest practices

Sandboxes, MCP Tunnels, Stainless: How Anthropic Is Pulling Ahead in Agentic AI

Three announcements in 48 hours. Anthropic brings execution in-house, connects agents to private systems without internet exposure, and strips OpenAI of its SDK generator. Analysis.

AnthropicClaudeMCPAI agentsStainlessSDKinfrastructure

Google just debunked GEO and llms.txt. Here's why I keep using them anyway.

On May 15, 2026, Google published its first official position on optimizing for generative AI. Verdict: no llms.txt, no structured data, no chunking. Plain SEO is enough. My nuanced opinion.

opinionSEOGEOllms.txtGoogleAI search

Claude Code vs Cursor vs GitHub Copilot: which one to pick in 2026?

An honest comparison of the 3 top AI coding tools in 2026. Claude Code, Cursor, GitHub Copilot: pricing, context window, agentic capabilities — and most importantly, which one wins for your profile.

Claude CodeCursorGitHub CopilotcomparisonAI toolsdevelopment

Claude for Small Business: how Anthropic just rattled Salesforce, Intuit and DocuSign

On May 13, 2026, Anthropic launched Claude for Small Business with QuickBooks, PayPal, HubSpot, Canva and Docusign. 3 named customer testimonials (Purity Coffee, Simple Modern, MidCentral Energy), 4 partner quotes, TechSy score 4.1/5, and a Wall Street that didn't take it well. Full breakdown.

ClaudeAnthropicSmall BusinessSMBCoworkQuickBooksHubSpotAI agents

The fading line: what Willison really admits about agentic engineering

Simon Willison coined the vibe coding / agentic engineering distinction. Then he admitted he no longer holds it himself. Addy Osmani, Amazon and Hacker News complete the picture.

opinionvibe codingagentic engineeringSimon Willisonsoftware engineering

2,541 GPT Conversations Analyzed: Why I Migrated Everything to Claude

I exported 2.5 years of ChatGPT data. 2,541 conversations, 13,250 responses, 119 Custom GPTs. Here's the full data-driven analysis of my migration to Claude Code.

Claude CodeChatGPTcase studydatabuild in public

Extended Thinking in Claude Code: When to Use It, When It's a Waste

Practical guide to Extended Thinking in Claude Code. Real refactoring and debugging cases, before/after benchmarks, and the pitfalls to avoid so you don't burn your tokens.

claudeclaude-codeextended-thinkingguidetokensai

Memory and Persistence in Claude Code: The Complete Guide (CLAUDE.md, MCP Memory, Sessions)

Everything you need to know about persistence in Claude Code. CLAUDE.md, memory files, MCP Memory, session storage: which mechanism to use and when.

claudeclaude-codememorypersistenceCLAUDE.mdMCPguideai

The Complete Anthropic Ecosystem Map (2026): SDKs, MCP, Agents, Open Source Tools

Complete guide to the Anthropic ecosystem in 2026. All SDKs, MCP servers, agent tools, and open source repos. What we actually use, what's noise.

anthropicclaudeecosystemSDKMCPagentsopen-sourceguideai

Don't Build Agents, Build Skills (the Lesson Anthropic Teaches in 16 Minutes)

Two Anthropic engineers explain why skills beat custom agents. Summary, analysis and practical lessons from Barry Zhang and Mahesh Murag's talk.

claudeanthropicagentsskillsclaude-codearchitectureai

Agent Skills: How Anthropic Created an Open Standard the Entire Industry Is Adopting (and Why It Changes Everything)

Anthropic's Agent Skills went from a Claude feature to an industry standard in 2 months. OpenAI is copying them, Microsoft is integrating them, and the ecosystem is exploding. Complete analysis of the standard, the strategy, and ground-truth feedback from a daily user.

Claude CodeAgent SkillsAnthropicopen standardAI agentsMCPproductivity

Claude Code Release Tracker: the Updates That Actually Matter (April 2026)

Monthly recap of Claude Code features that are actually game changers. Not raw changelogs: editorial analysis, real impact, and what it means for builders.

claudeclaude-codereleasechangelogupdateai

Claude Managed Agents: the Kill Switch Anthropic 'Forgot' (and Why It's Costing You)

Complete guide on Claude Managed Agents pitfalls: how tokens explode, why the kill switch is nowhere to be found, and the real safeguards to configure.

claudeanthropicagentsmanaged-agentstokensguideai

10 Custom Claude Code Agents Ready to Copy (With Full YAML)

10 production-ready Claude Code agents with complete YAML configs. Monitoring, daily recap, code review, business pipeline, tests, docs, debug, migration, content, security. Copy-paste and adapt.

claudeclaude-codeagentsskillsyamlguideai

Multimodal RAG in 30 minutes: Gemini Embedding 2 + Claude Code

Google releases its first natively multimodal embedding model. Text, images, videos, audio in the same vector database. Here's how to build a complete RAG pipeline with Claude Code.

technicalRAGGeminiClaude Codeembeddingsmultimodal

Claude Mythos and Project Glasswing: The Model Anthropic Refuses to Release (and Why It's a Turning Point)

Anthropic unveils Claude Mythos, its most powerful model: 93.9% on SWE-bench, 97.6% on math olympiads, thousands of zero-days discovered. But it won't be public. Full breakdown of Project Glasswing and what it changes.

ClaudeAnthropicMythoscybersecurityProject Glasswingbenchmarksnews

Claude Cowork: the complete guide to automating your daily work

Cowork turns Claude into a personal assistant that reads your files, manages your calendar and executes workflows. Complete guide: folders, instructions, connectors, skills, Dispatch and scheduled tasks.

guideClaudeCoworkautomationproductivityDispatch

Anthropic Cuts Off OpenClaw: Why Your AI Setup Should Never Depend on Anyone

Anthropic blocks Claude subscriptions via OpenClaw on April 4, 2026. Analysis of the facts, the strategic reasons, and what it means for developers building with Claude.

Claude CodeAnthropicOpenClawAI agentsMCPnews

Claude Code's Source Code Leaked: What We Found, What's False, and Why It Matters

512,000 lines of TypeScript exposed via npm. A fact-checked analysis of the leak, hidden features, the cache bug that multiplied costs by 20x, and what it means for users.

Claude CodeAnthropicsecurityleakanalysisnews

Claude and Claude Code FAQ: Everything You Need to Know in 2026

Answers to the most asked questions about Claude, Claude Code, Opus/Sonnet/Haiku models, pricing, MCP, and differences with ChatGPT and Gemini.

ClaudeClaude CodeFAQbeginnerguideAnthropic

15,000$, 8 Hours, Zero Hand-Typed Lines: Anatomy of an Anthropic Hackathon Win

Affaan Mustafa won the Anthropic x Forum Ventures hackathon using Claude Code. His everything-claude-code repo (117K+ stars) reveals an agent, skill, and memory setup that changes the game.

Claude Codehackathonagentsproductivityopen source

Pillar 3: security and hosting for AI builders (OWASP, pen test, scaling)

AI doesn't think like an attacker. OWASP Top 10, pen testing, hosting, load testing, scaling: the security checklist for builders coding with AI.

securityOWASPhostingdevopsbest practicesvibe codingscaling

Pillar 2: CI/CD and monitoring for solo builders (GitHub Actions, Sentry, uptime)

Manual deployment is the best way to ship broken code without knowing. CI/CD, Sentry, uptime monitoring: the minimum viable pipeline for solo builders.

CI/CDmonitoringbest practicessolo builderGitHub ActionsSentrydevops

Pillar 1: testing your AI code before shipping (TDD, unit tests, e2e)

AI generates code that works in the prompt. Not necessarily in production. Here's how to test code you didn't write: TDD, unit tests, integration, e2e.

testsTDDbest practicesClaude Codevibe codingsoftware engineering

The Checklist Before Shipping with AI: 3 pillars vibe coding ignores

You've been coding with AI for months. You ship fast. But do you test? Monitor? Secure? 3 pillars to go from vibe coding to software engineering.

vibe codingbest practicesopinionClaude Codesoftware engineeringmanifesto

How I built an AI copilot that runs my life as a solo builder

Complete architecture of a Claude Code system that manages my daily operations: automatic recaps, business tracking, content generation, competitive intelligence, and persistent memory. Everything is reproducible.

guideClaude Codeautomationsolo builderproductivityadvanced

The Claude Coder Paradox: why the tool that saves you time can waste it

You adopt Claude Code to move faster. Then you want to do everything, explore everything. The infinite multi-tasking trap, and how to escape it.

Claude Codeproductivitysolo builderopinionbest practices

Prompt engineering: 5 myths you need to stop believing

Viral 'perfect prompts' are theater. Breaking down prompt engineering myths with Anthropic's official recommendations and concrete examples.

prompt engineeringmythsAnthropicbest practicesClaude

Harrison Chase is right: coding agents are redefining development

Reacting to Harrison Chase's viral thread on how coding agents are restructuring product teams. As a solo builder with Claude Code, I live this thesis every day.

opinioncoding agentsClaude Codesolo builderHarrison Chase

Why Claude Code is a game changer in 2026

80,000+ GitHub stars, millions of developers, and yet almost no curated resource hub. Breaking down the tool that's redefining AI-assisted development.

opinionClaude CodeAIdevelopment