# Claude Ships Code > To ship with Claude Code, confirm `claude --version` (never pin it here), load a skill for the procedure you repeat, put gates in hooks, use MCP only for live context, and stop only on readable check evidence. Updated: 2026-09-19 Site: https://claudeshipscode.com Disclaimer: Independent educational publication. Not affiliated with Anthropic, Claude, or Claude Code. ## Signature - [The Terminal](https://claudeshipscode.com/terminal/): labeled synthetic Claude Code session. Version check (`claude --version`, unpinned), `/verify` skill as used in Code, hook/MCP beat, shipping check. Not a real transcript. Skill authoring and the skills directory live on Claude Has Skills: https://claudehasskills-com.pages.dev To ship with Claude Code, run a verification loop: confirm the binary with `claude --version`, load a skill for the procedure you repeat, put non-negotiable gates in hooks, reach for MCP only when the ticket lives outside the repo, and do not stop until a check produces evidence you can read. The Terminal plays that loop. For forensic log reading, use the Black Box on Claude Logs: https://claudelogs.com/black-box/ The specimen is synthetic. This site does not pin a Claude Code CLI version. Skill authoring belongs on Claude Has Skills, not here. ## Current Defaults (2026-09-19) - Claude Fable 5.1, model ID `claude-fable-5-1`, shipped 2026-09-01. Default effort is High in Claude Code, Medium on Claude.ai and Cowork. $10/$50 per MTok, 1M context, 128K max output, cache reads at 2.5% of base input. Source: https://www.anthropic.com/claude-fable-and-mythos-5-1 - Claude Opus 5, model ID `claude-opus-5`, shipped 2026-07-24. Recommended starting point for most workloads. $5/$25 per MTok, default effort high. Fast mode runs ~2.5x faster at 2x base price via the platform or Claude Code usage credits. Beta: mid-conversation tool changes without invalidating the prompt cache. Source: https://www.anthropic.com/news/claude-opus-5 - Claude Sonnet 5, model ID `claude-sonnet-5`. $2/$10 per MTok. The introductory rate became the standard rate and did not rise on 2026-09-01. - Claude Haiku 4.5, model ID `claude-haiku-4-5`. $1/$5 per MTok, 200K context. - Claude Opus 4.8 is legacy and is not the current Opus model. - Claude in Chrome is generally available on paid plans. This site treats it as an adjacent surface only; everyday how-to belongs to Claude Helps. ## Core Pages - [The Terminal](https://claudeshipscode.com/terminal/): labeled synthetic Claude Code session theater. Version check (`claude --version`, unpinned), `/verify` skill as used in Code, hook/MCP beat, shipping check. Not a real transcript. - [Claude Code Tutorial: Ship One Verified Change](https://claudeshipscode.com/start-here): first-session workflow, verification loop, prompt shape, review handoff. - [Claude Code Best Practices for Engineering Teams](https://claudeshipscode.com/best-practices): explore-plan-implement-review, context management, extension choice. - [CLAUDE.md and Settings](https://claudeshipscode.com/claude-md-settings): project instructions, settings scopes, local overrides. - [Claude Code settings.json Examples](https://claudeshipscode.com/settings-json): permissions, hooks blocks, command files, and settings scopes. - [Claude Code Hooks](https://claudeshipscode.com/hooks): deterministic lifecycle gates and safe hook design. - [Claude Code MCP](https://claudeshipscode.com/mcp): external systems, live context, protocol boundaries. - [Skills vs Subagents vs Plugins](https://claudeshipscode.com/skills-subagents): reusable procedures, isolated investigation, distribution. - [Claude Code in CI and GitHub Actions](https://claudeshipscode.com/ci-github-actions): official action, @claude workflows, CI guardrails. - [Security and Permissions](https://claudeshipscode.com/security-permissions): trust boundaries, secrets, MCP and hooks review. - [Agent SDK and Automation](https://claudeshipscode.com/agent-sdk-automation): CLI automation, SDK capabilities, structured outputs. - [Claude Code Changelog Watch](https://claudeshipscode.com/changelog): freshness surface for engineering-impact release notes. - [Primary Sources](https://claudeshipscode.com/sources): bibliography and claim map. ## Free Tools - [Claude Code Hooks Config Generator](https://claudeshipscode.com/tools/hooks-generator): client-side settings.json hooks block generator. - [Claude Code Permissions Profile Builder](https://claudeshipscode.com/tools/permissions-builder): client-side permissions.allow/ask/deny profile builder. - [Claude Code Slash Command Scaffolder](https://claudeshipscode.com/tools/command-scaffolder): client-side .claude/commands markdown generator. ## Primary Sources - Anthropic Claude Code docs: https://code.claude.com/docs/en/overview - Best practices: https://code.claude.com/docs/en/best-practices - Extension overview: https://code.claude.com/docs/en/features-overview - Permissions docs: https://code.claude.com/docs/en/permissions - Hooks reference: https://code.claude.com/docs/en/hooks - MCP docs: https://code.claude.com/docs/en/mcp - Skills docs: https://code.claude.com/docs/en/skills - Skills and slash commands: https://code.claude.com/docs/en/slash-commands - Subagents docs: https://code.claude.com/docs/en/sub-agents - GitHub Actions docs: https://code.claude.com/docs/en/github-actions - Official action repo: https://github.com/anthropics/claude-code-action - Agent SDK: https://code.claude.com/docs/en/agent-sdk/overview - Changelog: https://code.claude.com/docs/en/changelog - Models overview: https://docs.anthropic.com/en/docs/about-claude/models - Choosing a model: https://platform.claude.com/docs/en/about-claude/models/choosing-a-model - Fable 5.1 announcement: https://www.anthropic.com/claude-fable-and-mythos-5-1 - Fable 5.1 platform overview: https://platform.claude.com/docs/en/models/fable-5-1/overview - Opus 5 announcement: https://www.anthropic.com/news/claude-opus-5 - Claude in Chrome GA: https://claude.com/blog/claude-in-chrome-generally-available Optional: [Full LLM context](https://claudeshipscode.com/llms-full.txt)