claude-code
7 posts
When a coding agent underperforms, the reflex is a stronger model. On bounded tasks the harness moves the score at least as much; a rule for which lever to pull.
ai-agents · ai-tools · llm +3
Devcontainers, Codespaces and AWS Lambda MicroVMs as homes for a coding agent: what each rung adds, what it costs, and when moving the agent off the laptop pays off.
lambda · claude-code · ai-tools +5
Agents made code-writing essentially free, but judgment about when and how much to use them is still entirely yours. An Aristotelian frame to separate the two skills.
ai-tools · claude-code · ai-agents +3
A practical repo layout that keeps Claude Code, Codex, Copilot, Cursor, and OpenCode reading the same rules, with honest notes on where portability breaks.
ai-tools · claude-code · github-copilot +3
A framework for six levels of AI assistance in software, from code review to vibe coding, with guidance on when to dial AI help up or down.
ai-tools · code-quality · productivity +4
How GitHub's Spec Kit turns loose AI code generation into structured, maintainable output through a four-phase specify-plan-tasks-implement loop.
ci-cd · ai-tools · code-quality +4
A comprehensive guide to Claude Code, AI agents, and Model Context Protocol servers that transforms developers from basic users to power users
claude-code · mcp · ai-tools +5