AI Coding

Hands-on tutorials for coding with AI. Claude Code, MCP, agents, prompts.

• 
  10 min · Advanced

  Claude Code hooks: automate build, format, and notifications without relying on the model

  Hooks move some guarantees out of the prompt. You can format, block a command, run a check, or notify without hoping the agent remembers.

  Claude CodeHooks
• 
  10 min · Intermediate

  Build an AI coding editorial agent with Codex without generating hollow content

  A good editorial agent should not produce more text. It should choose better topics, verify facts, and reject weak angles.

  CodexSkillsAstro Content
• 
  11 min · Intermediate

  Codex: AGENTS.md, skills, subagents, what does what?

  Codex has several instruction surfaces. The trap is putting everything in AGENTS.md. Here's when to use repo guidance, a skill, a custom agent, or a subagent.

  CodexAGENTS.mdSkills
• 
  10 min · Intermediate

  Copilot cloud agent: from issue to PR without losing control

  A good Copilot cloud agent workflow starts before assignment. A fuzzy issue creates a fuzzy PR. Here's how to scope, constrain, and review.

  GitHub CopilotCloud agentGitHub Issues
• 
  9 min · Intermediate

  Copilot Code Review: when to click, when to keep a local checklist

  AI review consumes quota or credits depending on your plan. It can help, but it does not replace your own checklist for business logic, security, and migrations.

  GitHub CopilotCode review
• 
  8 min · Intermediate

  Cursor iOS: drive a mobile agent without merging blindly

  Cursor iOS makes agents reachable away from the laptop. The real workflow is not coding on mobile, but launching, monitoring, reviewing, and deferring risky decisions.

  CursoriOSAgents
• 
  9 min · Intermediate

  Useful MCP for AI coding: logs, issues, and docs only when it removes copy-paste

  MCP is not magic. It becomes useful when it replaces repetitive manual transfer between a real tool and your coding agent.

  MCPClaude CodeCodex
• 
  9 min · Advanced

  Codex subagents: clean fan-out without polluting context

  Subagents help when a task really splits apart. Otherwise they add cost, latency, and noise. Here's the grid for using them cleanly.

  CodexSubagents
• 
  15 min · Intermediate

  Claude Opus 4.7 in Claude Code: the guide to not burning your tokens

  Opus 4.7 is a leap forward for agentic coding, but it'll eat 30–60% more tokens if you keep driving it like Opus 4.6. Effort levels, adaptive thinking, default behaviors — here's how to actually adapt your workflow.

  Claude CodeClaude Opus 4.7