#developer-workflow

Claude Code shipped four tagged releases in roughly 50 hours, one switching a tool family off by default, reversibly. Running a team on a shifting harness.

Text and DOM output lets an agent close its own loop. 3D, video and physics do not. A verifiability ladder for knowing when you are still the checker.

An agent stops when its work looks done. Write the completion bar instead of asking for effort: a named check, required evidence, iterate until it passes.

Claude Code documents a --worktree flag, Copilot CLI's /worktree is experimental, Muse Code's subagent isolation is reported. Why agents need it, and its cost.

Running Kimi K3 in Kimi Code: the Moderato plan unlocks 256K context, Allegretto the full 1M, and cache discipline — not knobs — controls your real cost.

Fable 5's retrained safety classifier blocks the reported jailbreak 99% of the time but flags more real code. Coding trade-offs, plus fixes dev teams can use.

Hands-on deep dive into Anthropic's MCP Inspector — local test loop, tool-call visualization, manifest validation, transport testing, debugging patterns.

Build a Claude Skill end-to-end: SKILL.md, scripts, references, allowed-tools, and ship as a project or user slash command. Working example included.

Build a CLI that reads your staged diff and generates conventional-commits messages with Claude Haiku 4.5. Bash hooks, prompt design, CI wiring inside.

Build a working MCP server in TypeScript: tool registration, JSON-RPC handlers, Claude Desktop integration, and a publishable npm package end-to-end.

Build a working Claude Code subagent from .claude/agents/ markdown to first invocation. Tool restrictions, system prompt design, orchestration patterns.

Claude Code Remote Control enables session handoff between local terminals and remote servers. Setup, security model, and use cases for AI development.

Automate code reviews with AI. Cursor, GitHub Bugbot, Claude Code integration. Complete workflow guide with best practices.