Supervise Claude Code Sessions with Cc Haha
Use Cc Haha as a local desktop workspace for Claude Code sessions, worktree launches, diff review, permission approvals, Computer Use, remote access, and scheduled agent tasks.
npx skills add agentskillexchange/skills --skill supervise-claude-code-sessions-with-cc-haha
Use Cc Haha to run Claude Code work from a local desktop workspace: select a project, launch a session against the current branch or an isolated worktree, monitor tool calls and task activity, review generated file diffs, approve risky commands, and continue or inspect work from remote access surfaces when needed.
What this skill actually does
What the agent/operator does
When to invoke
Invoke this skill when a user needs supervised Claude Code work that is easier to review than a terminal transcript: multi-project sessions, worktree isolation, visible diffs, permission gates, model setup, Computer Use authorization, background activity tracking, or mobile/IM continuation for a running task.
Scope boundary
This is not a generic desktop app card or a replacement for normal Claude Code prompting. Keep the workflow bounded to supervising Claude Code sessions, launch context, approval mode, diff review, and session continuity through Cc Haha’s documented workspace controls.
Inputs and prerequisites: Cc Haha desktop app, Claude Code, a configured model provider or local model endpoint, repository access, optional Computer Use and remote access integrations.
Setup notes: Download the macOS, Windows, or Linux desktop installer from the GitHub Releases page, launch Cc Haha, configure the model provider/API key/default model in Settings, then open a project and start a Claude Code session from the workspace.
Source and verification boundary: use https://cchaha.ai as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Claude Code workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.