tmux plugins

Orchestrate parallel agent workflows across isolated tmux-managed git worktrees: spawn agents, monitor their status as tmux window names update in real time, and merge worktree branches with conflict resolution.

Analyze codebases for semantically duplicate functions to guide refactoring, run interactive CLI tools via tmux detached sessions when piping fails, discover and invoke MCP server capabilities on-demand for testing, and spin up headless Windows 11 VMs in Docker with SSH access for cross-platform development.

Spawn any third-party LLM with an Anthropic-compatible API (DeepSeek, GLM, Kimi, Qwen, MiniMax) as Claude Code teammates in tmux or as one-shot subagents — multi-turn collaboration, fan-out tasks, or scripted multi-phase workflows. Provider API keys are isolated via apiKeyHelper, never exposed in env/argv/history.

Run multi-AI roundtable discussions (Codex, Claude, Gemini) in tmux with @-mention chat, then promote consensus to an execute-mode swarm where a supervisor dispatches tasks to role-based agents (engineers, reviewers). Includes isolated LLM voting with multi-round debate and file injection for independent judgments.

Automate Claude Code workflows: scaffold greenfield projects via interview-driven decisions and markdown specs, generate layered implementation plans with approvals, audit contexts/skills/plugins for quality and drift, orchestrate optimal agent patterns for complex tasks, assess harness maturity across axes, analyze session execution patterns, and QA web/desktop/CLI apps with health reports.

Orchestrate Claude Code sessions as a Kanban board with agent lifecycle hooks, role-based task assignment, and automated verification of web pages, CLI tools, APIs, databases, and libraries, plus fuzzy exploration and plan decomposition into worker-ready tickets.

Perform macOS security audits and exploitation including sandbox escape, TCC bypass, dylib injection, XPC attacks, kernel exploitation, and threat intelligence queries. Also debug interactive terminal-based tests using tmux.

Delegate code tasks like review, planning, and analysis to token-efficient Haiku subagents that run OpenCode headlessly, with prompt templates and result validation, saving 70-80% tokens.

Orchestrate a structured per-feature development lifecycle using autonomous agents: design, implement, validate, and wrap features in isolated git worktrees with automated feedback loops and artifact validation.

Build Pi coding agents with grounded skills: deploy vLLM GPU pods on DataCrunch/RunPod, integrate via RPC/Node SDK in TypeScript apps, embed Lit web UIs and TUI components, author extensions/packages, customize prompts/behaviors, manage CLI workspaces with tmux, and deploy Dockerized Slack bots.

Drive Navaris CLI with AI to fully manage isolated dev sandboxes: create/start/stop/destroy using Incus or Firecracker backends, exec one-shot commands with env/timeout, attach persistent tmux terminals, create/restore snapshots, promote/register images, and handle async operations.

Track Claude Code's activity in real-time via status symbols (○ for session start, ✻ for prompts/tool use, ✓ for stop, ⌫ for compact) shown in tmux window and pane titles. Persists across panes, supports /rename matcher, and runs via shell commands for seamless terminal awareness without context switching.

Launch and play Dwarf Fortress-inspired ASCII simulations in tmux terminals, with Claude as your interactive fortress overseer. Generate dramatic fortress names to embark on new worlds or use an arrow-navigated picker to resume, delete, or manage saved games via CLI commands powered by a local stdio server.

Run fully autonomous AI coding sessions that generate step-by-step plans, execute tasks with self-critique and best practices, enforce file policies to prevent clutter, block risky commands via safety guards, manage tmux for isolated testing, and export refined plans.

Orchestrate multi-agent development teams in Claude Code: plan complex tasks, delegate to specialized agents for coding, searching, reviewing, and debugging, then gate-merge changes through parallel quality and security checks. Manage persistent project sessions from Telegram or Slack.

Run autonomous, personalized research loops on any topic to generate quality-gated reports tailored to your projects and domains. Use presets like technique-scout or cross-domain; set it via tmux session, walk away, and return to results. Stop anytime.

Embed interactive UI elements like confirmations, progress trackers, code diffs, data tables, and charts into Claude Code agent workflows using awb CLI. Launch with `awb run --title "Title" <component> [args]` to send interactions and get an ID, then pause execution with `awb wait <id>` until responding in the awb UI playground opened via `awb ui`. Ideal for safe destructive actions, multi-step tasks, git reviews, and structured outputs.

Automate full GitHub project lifecycle from PRD issues to merged code: create structured PRDs, decompose into epics/tasks, execute parallel implementations in git worktrees via agent teams, perform code reviews, generate/run Playwright QA tests, fix CI failures, resolve conflicts.

Automate Claude Code development with a multi-phase, contract-gated team of 7 specialized agents for planning, verification, monitoring, documentation, and orchestration, plus security hooks against dangerous commands and secrets.

Audit dotfiles for security vulnerabilities, performance bottlenecks, and adherence to modern CLI best practices, then receive a scored health report with actionable recommendations for Zsh, Tmux, Neovim, and Git configurations.

Label a GitHub issue to trigger Zapat's autonomous pipeline powered by Claude Code, which builds, tests, and delivers a merge-ready PR. Setup interactive agent teams, monitor pipeline health with auto-fixes, and add repos for ongoing automation.

Orchestrate multi-model AI coding workflows with prompt management, git worktree isolation, and agent-team coordination for complex multi-step tasks across Claude, Codex, and Gemini.

Automate specs-to-code workflows with an autonomous loop that verifies outputs via multi-model cross-checking. Spawn parallel agent teams (explorer, coder, architect) in tmux workspaces with guard hooks enforcing safety, conventions, and scope.

Start, stop, restart, and monitor services running in dedicated TMUX panes without cluttering your terminal view. Automatically activates when running inside a TMUX session.

Manage interactive PTY sessions for local shells, SSH, serial ports, and persistent remote terminals from AI agents, enabling server command execution, device interaction, and log monitoring.

Coordinate multiple Claude Code agents for epic planning, task execution, and inter-agent communication via tmux and MCP, with automated iOS TestFlight deployment using Fastlane and GitHub Actions.

Stream Twitch, YouTube, and Kick audio via mpv inside Claude Code while viewing live chat in a tmux pane, with controls for play/pause, volume, mute, and layout switching.

Convert static web or Next.js projects into ArcBlock blocklets with a provided DID, automate git branch detection and PR creation following repo conventions, set up blocklet dev environments by cloning repos and installing deps, guide onboarding for new developers, and streamline releases via version bumps and changelog updates.