Everyday AI coding workflows

Give your coding assistant clear instructions, share the right project context, investigate errors and check terminal commands before running them.

Turn a coding task into a repeatable workflow: define the expected result, supply relevant files, investigate the cause of failures and verify changes with focused tests.

Markdown

• Claude Code shortcuts list: 11 practical tips and commandsUse this Claude Code shortcuts cheat sheet for Shift+Tab, Esc Esc, /compact and /cost. Claude Code shortcuts and commands help manage modes, context and usage.
• Claude Code AGENTS.md vs CLAUDE.md: one shared instruction fileCreate a Claude Code AGENTS.md file, choose a Claude Code AGENTS.md symlink or import, and share rules with Codex and Cursor.
• What Is AGENTS.md and How to Write Your Own in 30 MinutesAGENTS.md format, support in AI tools, a ready-made template, rules, and mistakes. How to coordinate AGENTS.md with CLAUDE.md and other instruction files.
• Context Engineering vs Prompt Engineering: Claude Code GuideWhat context engineering is and how it differs from prompt engineering. The Anthropic canon, 7 mistakes, and a minimal setup for Claude Code.
• Claude Code 2.1.268: What Was Fixed in deny Rules and MCPClaude Code 2.1.268–2.1.269 fixed four permission issues and a secret leak in /mcp. Who is affected, how to update, and how to check your rules.
• Claude Code Now Reads AGENTS.md: What Changed in 2.1.277On September 18, 2026, Claude Code 2.1.277 gained direct support for AGENTS.md: when the file is loaded, how to choose one of 4 modes, and what to do about imports.
• Auto Mode in Claude Code: Free Security Checks and the “this session isn't eligible” NoticeStarting with version 2.1.278, Claude Code does not charge for auto mode security checks on the server. What “this session isn't eligible” means and how to check the status.
• Claude Code: 30-minute background command limitClaude Code stops a background command after 30 minutes in unattended runs: -p, Agent SDK, CI, cloud. How to raise the limit and what to do after a command stops.
• How to disable Claude Code auto memory in one projectClaude Code auto memory in a single project is turned off with the line autoMemoryEnabled: false in .claude/settings.json. How is this better than /memory, and how do you clean up notes?
• Claude Code deletes history after 30 days: how to save itWhere Claude Code sessions are stored, how to configure cleanupPeriodDays, and how to organize an external archive using rsync, launchd, or cron.
• How to download Claude Desktop on your computer and phone, and where to startOfficial download and installation of Claude on Windows, Mac, Linux, and phone: requirements, synchronization, and the differences between Chat, Cowork, and Code.
• Claude Desktop on Windows: file is in use after an updateWhy Claude Desktop on Windows reports that the file is in use by another program, how to restore startup, and what to send Anthropic Support.
• CLAUDE.md After Opus 5: Why Long Instructions Now Get in Claude’s WayClaude Opus 5 is smarter, and old, lengthy rules in CLAUDE.md get in its way. What to keep, what to move to skills, and how to enable automatic memory. Includes a ready-to-use prompt.
• Claude Memory: Memory, CLAUDE.md, and Auto MemoryThree Claude memory systems: where they work, how to manage the settings, what to store, and how to avoid five common mistakes.
• How to upload a file to Claude so it can read itClaude in the browser accepts PDFs, spreadsheets, images, and code. All the formats, how large a file can be and how many fit in a chat, and what to do when an upload fails.
• Voice mode in Codex: how to enable /voice and what it can doVoice conversations with an agent appeared in Codex CLI 0.155. How to update, enable /voice via /experimental, use the mute and stop commands, and understand the experiment’s limitations.
• Free Voice Input: Local Whisper and Ready-Made AppsWhisper, Wispr Flow, and superwhisper: voice input setup, computer requirements, privacy, and troubleshooting.
• Four prompts to counter sycophancy in Claude CodeFacts, precedents, a plan, and a self-check: a four-step cycle inside Plan Mode for complex tasks and unclear errors.
• How to Set Up CLAUDE.md in 2026: Template and 6 RulesA ready-to-use CLAUDE.md template, 6 rules for keeping the file alive, and an analysis of a working file from an active business. Includes official Anthropic quotations.
• How to Configure Claude for Yourself: Profile, Memory, and ProjectsFour layers for configuring Claude: profile instructions, memory, projects, and CLAUDE.md. A complete interview prompt, a result check, and a review of the rules.
• How to update Claude Code to the latest version in 2026How to check the Claude Code version, update it with the claude update command, install the latest build via Homebrew or WinGet, and what to do if it won’t update.
• How to undo changes in Claude Code: /rewind and gitRewind modes, checkpointing limitations, git checkout, stash, reflog, and protection against destructive commands. Five ways to restore a state.
• Claude Code Recovery: Safe Mode, /cd, and FallbackDiagnosing Claude Code with safe mode, the /cd command, fallback models, and settings recovery. The order of checks and original limitations.
• How to install Claude Code from the official websiteFor a Claude Code install terminal users can choose a native installer. The Claude Code install npm method requires Node.js 22 or later.
• How to Run Qwen Code in the Terminal and Set Up Edit ModeInstalling Qwen Code, launching it in a project, connecting a model through auth, Plan, Auto-Edit, Auto, and YOLO modes, a local model, and project protection.
• How to Turn PDF and Word Files into Text for AI: Microsoft’s MarkItDownIn-depth review: How to turn PDF and Word files into text for AI: Microsoft’s MarkItDown.
• Response styles in Claude Code: how to switch and create your ownWhat output styles in Claude Code are and how they differ from CLAUDE.md: four built-in styles, switching via /config, and how to write your own.
• PDFs in Claude Code: Conversion, Tokens, and Preserving StructureMarkItDown, PyMuPDF4LLM, MCP, complete conversion scripts, PDF limits, and cases when visual analysis matters more than token savings.
• Plan Mode in Claude Code: Step-by-Step Guide and 4 Phases in 2026Plan Mode in Claude Code: what it is, how to enable it with Shift+Tab, Anthropic’s 4 phases, ready-to-use prompt templates, and the /goal command for autonomous overnight tasks.
• Why AI tools become frustrating: a practical learning mapAn author’s four-stage learning metaphor, practical workflow signs, and a three-pillar experiment for making progress on one AI-assisted project.
• Claude Code can’t see project files: why and what to do in 2026Why Claude Code can’t find project files: it was started from the wrong folder, the file is in a subfolder, or it’s in .gitignore. How to grant access with add-dir, cd, and settings.
• Why Claude Code rereads the entire project and how to fix itClaude Code doesn’t index code and rereads the project from scratch, burning tokens. How to give it persistent code memory via MCP and reduce usage.
• Why Claude Does the Wrong Thing: 3 PromptsClaude writes the wrong thing because it rushes into code without understanding your intent. I explain why and give you 3 ready-to-use prompts to make it ask questions first.
• Why an AI agent says “done” when the work isn’t finishedReward hacking, completion criteria, independent acceptance, and two ready-to-use prompts for checking the actual result.
• Qwen Code stopped working: why and what to do in 2026I explain why Qwen stopped working: free access was closed, chat and terminal got confused, and there was a 401 error. How to get the agent working again without a card.
• How to Set Up Spec Kit in Claude Code in 2026: Specifications Instead of PromptsHow to set up GitHub Spec Kit in Claude Code: 4 phases—Specify, Plan, Tasks, Implement. Here’s what replaces improvised vibe coding.