skills/tmux-workflow/SKILL.md
Use when running multiple parallel agents, monitoring long-running background tasks, or requiring a multi-pane observability dashboard for parallel implementation plans.
npx skillsauth add hangsiahong/koompi-agent-bundle tmux-workflowInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
A tmux-based Command Center for managing parallel agent execution, background observability, and automatic code indexing.
bash scripts/tmux-manager.sh init "ProjectName" "/path/to/project" --auto-index
The --auto-index flag automatically starts the code indexer in the Monitor pane. It will clean orphans, index new/changed files, and show status — all hands-free.
bash scripts/tmux-manager.sh init "ProjectName" "/path/to/project"
bash scripts/tmux-manager.sh reindex "ProjectName" "/path/to/project"
Sends the auto-index command to Pane 1. Use this after major code changes during a session.
bash scripts/tmux-manager.sh status "ProjectName" "/path/to/project"
Shows tracked file count, new/changed/deleted counts, and DB size — without starting a full index.
bash scripts/tmux-manager.sh add-worker "ProjectName" "Worker Name"
The tmux workflow is tightly integrated with the code-intelligence skill:
--auto-index: Automatically keeps the LanceDB index freshreindex command: Trigger re-indexing from any point in the sessionstatus command: Quick health check without leaving tmuxThe indexer tracks file changes via mtime fingerprints, so re-indexing only processes what actually changed — fast and cheap.
The tmux-manager.sh script handles session naming, pane splitting, layout optimization (usually tiled or main-vertical), and delegates to the code-intelligence indexer for all indexing operations.
testing
Use when creating new skills, editing existing skills, or verifying skills work before deployment
development
Use when you have a spec or requirements for a multi-step task, before touching code
data-ai
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
tools
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions