Skip to content

Tag

Vibe Coding & Claude Code

Tutorials, guides, and deep dives on claude code for AI-assisted development.

39 articles

pulse·intermediate·13 min

Claude Code Rewind Recovers Your Session After Slash Clear

Claude Code v2.1.191 adds /rewind to restore sessions after /clear, while v2.1.193 adds classifyAllShell for complete auto-mode shell command coverage.

Read
pulse·intermediate·12 min

Claude Tag Turns AI Coding From Solo Sport to Team Play

Claude Tag puts @Claude in Slack as a shared team agent with persistent memory and ambient mode. Anthropic says 65% of product code now comes from it.

Read
pulse·intermediate·11 min

Claude Code v2.1.186 Fixes Silent Background Agent Failures

Claude Code v2.1.186 (June 22) ends silent background agent failures. Permission prompts now surface to your main session. Adds claude mcp login for SSH.

Read
pulse·intermediate·8 min

Claude Design Learns Your Brand and Hands Off to Claude Code

Claude Design's June 2026 update adds design system imports from GitHub, /design-sync in Claude Code, and two-way sync that ends the mockup rebuild loop.

Read
pulse·intermediate·15 min

Claude Fable 5 Offline Since June 12 and What Comes Next

Fable 5 and Mythos 5 went offline June 12 under a US government export control directive. Claude Code defaults to Opus 4.8. ID verification starts July 8.

Read
pulse·intermediate·12 min

Claude Code Artifacts Turn Sessions Into Live Shared Pages

Claude Code Artifacts, launched June 18, lets Team and Enterprise plans turn any coding session into a live shareable HTML page that updates as Claude works.

Read
pulse·intermediate·11 min

Claude Code Permission Rules Now Target Specific Subagent Models

Claude Code v2.1.178 ships Tool(param:value) permission rules so teams can block Opus subagents, risky bash patterns, and MCP writes with one wildcard entry.

Read
pulse·intermediate·12 min

Claude Code Now Lets Sub-Agents Spawn Their Own Sub-Agents

Claude Code v2.1.172 ships nested sub-agents up to 5 levels deep, letting agents delegate to children. Keeps each context lean for complex multi-step tasks.

Read
pulse·intermediate·12 min

Claude Fable 5 Is Live in Claude Code With 80 Percent SWE Score

Claude Fable 5 is in Claude Code v2.1.170. Posts 80.3% on SWE-Bench Pro, leading Opus 4.8 by 11 points. Use /model fable, free on subscriptions to June 22.

Read
pulse·intermediate·11 min

Claude Code Safe Mode Lands in v2.1.169 With 30 Bug Fixes

Claude Code v2.1.169 ships --safe-mode to disable all customizations for debugging, a /cd command for session directory moves, and 30-plus reliability fixes.

Read
pulse·intermediate·11 min

Claude Code Adds Fallback Models and Tightens Agent Security

Claude Code v2.1.166 ships fallback models to survive API overloads automatically, and closes a SendMessage privilege gap affecting multi-agent orchestration.

Read
pulse·intermediate·10 min

Why Microsoft and Uber Hit the Claude Code Budget Ceiling

Microsoft canceled Claude Code licenses across its Experiences and Devices group. Uber burned its 2026 AI budget in four months. Lessons for every builder.

Read
pulse·intermediate·11 min

Claude Code Dynamic Workflows Run Up to 1000 Parallel Agents

Claude Code ships dynamic workflows alongside Opus 4.8. Claude writes orchestration scripts that fan work to up to 1000 subagents, 16 concurrent, resumable.

Read
pulse·intermediate·11 min

Claude Code Security Plugin Scans for 25 Vulnerability Classes

Claude Code's new security plugin ships free on May 27. It scans edits and commits for 25 vulnerability classes using a fresh Opus 4.7 reviewer on all plans.

Read
pulse·intermediate·11 min

Why Claude Agents Now Run Inside Your Private Network

Claude Managed Agents shipped self-hosted sandboxes and MCP tunnels. Your code stays in your infrastructure and private APIs need no public firewall holes.

Read
pulse·intermediate·11 min

Claude Mythos 1 Is Coming to Claude Code What to Expect

Claude Mythos 1 appeared as a toggle in Claude Code on May 23 before vanishing. What its 93.9% SWE-bench score means for vibecoders and when access is expected.

Read
pulse·intermediate·11 min

Mini Shai-Hulud npm Attack Targets Your Claude Code Config

Mini Shai-Hulud hijacked Claude Code's SessionStart hook as a persistence vector in May 2026. If you ran npm install recently, here is your cleanup checklist.

Read
tools·intermediate·16 min

How to Structure Claude Code Skills, the Authoring Guide

The folder layout, SKILL.md frontmatter, description field, and decision tree for authoring Claude Code skills that activate when you actually need them.

Read
tools·intermediate·22 min

Claude Code Slash Commands, the Complete Reference for 2026

A complete reference to Claude Code slash commands, including all 70+ built-in commands, cost warnings, custom command authoring, and community resources.

Read
tools·advanced·12 min

Claude Code MCP Servers, Postgres, GitHub, and Custom in 2026

Claude Code MCP servers tutorial. Install Postgres and GitHub MCPs in 5 minutes, then build your own. With working configs and a complete custom-server example.

Read
build·intermediate·12 min

Claude Code Stop Hooks Setup Walkthrough, 30 Minutes

Claude Code stop hooks tutorial. Configure .claude/settings.json, write the hook script, and let Claude run autonomously for hours. Working example included.

Read
tools·intermediate·13 min

CLAUDE.md File Format, Sections, and Real Examples

CLAUDE.md file format guide. The 5 sections every project needs, the 3-tier hierarchy Boris uses, plus templates for Next.js, Python, and monorepos.

Read
tools·intermediate·11 min

How the Claude Code Team Actually Uses Claude Code

The Claude Code team's own playbook: plan mode first, CLAUDE.md as memory, verification as the #1 lever. Named quotes from Boris, Cat, Thariq, Sid.

Read
build·intermediate·13 min

Building a verify-app Skill for Claude Code, Worked Example

Build a /verify-app skill for Claude Code in 30 minutes. The folder layout, SKILL.md contents, and the scripts Boris and Cat actually recommend.

Read
pulse·beginner·9 min

Inside Anthropic, the Team That Ships 90% of Code With Claude

90% of Anthropic's code is Claude-written. 150% productivity gain. 67% more PRs per dev. The receipts, with named sources from Boris, Cat, and Sid.

Read
build·intermediate·11 min

Plan Mode Is the 2-3x Cheat Code Most Claude Code Users Skip

Plan mode doubles or triples Claude Code's success rate. Here's how Boris, Cat, Sid, and Thariq use it every day, with the exact shortcuts and patterns.

Read
build·intermediate·13 min

Verification Beats Prompting, Boris Cherny's Top Claude Code Tip

The biggest quality lever in Claude Code is letting Claude verify its own work. Boris, Cat, and Thariq all converge on this. Here's how to set it up.

Read
build·advanced·11 min

Multi-Claude, How Anthropic Engineers Run 5+ Agents in Parallel

Boris runs 5 terminal Claudes plus 5-10 web sessions plus mobile. Sholto multiplexes. Here's the parallelism workflow Anthropic engineers actually use.

Read
build·advanced·12 min

Loops, Stop Hooks, and Scheduled Agents in Claude Code

Boris runs dozens of cron-scheduled agents for PR maintenance and CI health. Cat's Routines lets you schedule Claude Code itself. Here's the pattern.

Read
tools·advanced·11 min

Skills, Slash Commands, and Subagents in Claude Code

Boris turns daily tasks into slash commands. Thariq says skills are folders not files. Subagents exist for parallelism. How they work together.

Read
pulse·intermediate·11 min

Surfing the Models, an Outsider View on Claude Code

Vaibhav Gupta calls Claude Code's team a compiler. Dex Horthy says exhaust the first while loop first. The contrarian view that matches the data.

Read
tools·intermediate·12 min

Claude Code Memory, How /memory and CLAUDE.md Actually Work

Claude Code memory has three layers: user-level, project-level, and Memory.md scratchpad. Here's how Boris, Cat, and Thariq actually use each one.

Read
tools·intermediate·11 min

Claude Code vs Cursor vs Windsurf in 2026 and Which One Wins

Claude Code vs Cursor vs Windsurf compared for 2026. Detailed breakdown of reasoning quality, pricing, workflow, and which tool fits developers, founders, and teams.

Read
tools·intermediate·11 min

Devin vs Claude Code in 2026, Which AI Coder Wins

Devin vs Claude Code in 2026. Honest verdict, pricing, task suitability, and when autonomous Devin beats collaborative Claude Code (and vice versa).

Read
tools·intermediate·12 min

25 Claude Code Tips That Most People Miss in 2026

Most Claude Code users miss these 25 power tips. From CLAUDE.md setup to parallel subagents, hooks, and MCP servers, these features save hours weekly.

Read
tools·intermediate·10 min

Claude Code vs Copilot and the Terminal vs Editor Paradigm

Claude Code vs GitHub Copilot compared. Terminal agent vs editor autocomplete, two paradigms for AI coding with different strengths for different workflows.

Read
build·intermediate·11 min

How to Prompt Differently for Cursor, Claude Code, and Copilot

Prompt differently for each AI tool. Learn the optimal prompting style for Cursor Composer, Claude Code agent mode, and GitHub Copilot inline suggestions.

Read
tools·intermediate·12 min

Cursor vs Claude Code and When to Use Which in 2026

Cursor vs Claude Code comes down to GUI vs terminal. Learn which AI coding tool fits your workflow, with honest pricing and feature comparison for 2026.

Read
tools·intermediate·13 min

Claude Code, the Complete Guide to AI Coding in Terminal

Claude Code is Anthropic's terminal-based AI coding tool. Learn setup, agentic workflows, CLAUDE.md files, and when CLI-first development beats an IDE.

Read

The Tuesday Shipping Report

Every Tuesday, one focused email:

  • - The tool or technique that's actually working right now
  • - A real problem from the community (and how to solve it)
  • - What changed this week in the vibe coding landscape

Read by 1,000+ founders, developers, and creators building with AI. Free forever. No spam.