How i handle complex tasks with Claude Code

#ai #programming #claudecode #development

Every big task I have that needs attention from several repos, I like to set up a fresh isolated folder for Claude with everything it needs. Searching, fetching, cloning the right repos every single time - it's repetitive and annoying. That's why I built claude-clone.

The Real Problem With Multi-Repo Work

When Claude Code opens a folder, it sees everything in it. Clone multiple repos into the same workspace and Claude understands how they all relate — your API and frontend, your shared lib and the services using it. The context quality jumps.

But nobody talks about the setup cost. You have to manually find each repo, clone it, and assemble the workspace every single time.

One Command Instead

claude-clone create my-feature

Choose your desired repos for the task:

That's it. It:

Fetches all your GitHub repos (org or personal)
Shows a searchable list, space to select
Clones everything you picked in parallel
Writes a CLAUDE.md describing the workspace
Launches Claude with full context across all repos

No config files to edit. No paths to remember. No repetitive setup.

Save Repo Groups as Presets

Working on the same stack repeatedly? Save it once:

claude-clone preset save backend
claude-clone create my-feature --preset backend

Works for Orgs and Personal Repos

claude-clone create my-feature --org acme   # org repos
claude-clone create my-feature              # your own repos
claude-clone open my-feature               # reopen later