One env var gives Claude Code separate config profiles
A developer posted on Hacker News about running multiple independent Claude Code setups side by side. CLAUDE_CONFIG_DIR points at ~/.claude by default; create a different directory, launch claude with the variable set to it, and you get a clean instance with its own skills, plugins, and MCP connections. He wraps each one in an alias in ~/.bashrc or ~/.zshrc.
His own setup runs an experimental build loaded with community plugins (caveman, i-have-adhd, context-mode) alongside a separate build tied to a different account.
For managing and sharing these profiles he points to his own tool, Claude Playbooks: a single static binary with no dependencies, MIT licensed, source on GitHub.
Up next
agent-mesh lets coding agents talk to each other over MCP
agent-mesh is an open source MCP server that lets six coding agents (Claude Code, Codex, opencode, Gemini, Grok, cursor-agent) open each other’s sessions, send prompts, and read the replies. Handoff depth defaults to 4, with a 300-second per-turn timeout. MIT licensed.
Related stories
- Intent-lens spells out Claude Code calls before they run
- Codedict cuts Claude Code tokens 63%, but cost only 26%
- claude-files-pane shows Claude Code’s file reads live
- lcc cut one Claude Code session 49.5% without a summary
- catnap keeps Claude Code from compacting mid-task
- Script checks whether your network breaks Claude API uploads
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.