Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By lobu-ai
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.
npx claudepluginhub joshuarweaver/cascade-ai-ml-agents-misc-1 --plugin lobu-ai-agent-workbenchYour Command Center for Claude Code.
https://github.com/user-attachments/assets/b7882c4f-c2bc-43e6-97bb-eb9353d06f28
Running multiple Claude sessions? Agent Workbench gives you a single playground to monitor them all. See which agents are running, thinking, waiting, or idle. Respond to interactions when you're ready—agents wait patiently.
awb ui
Claude Code's built-in AskUserQuestion tool blocks execution until you respond. One question at a time. If Claude needs approval mid-task, everything stops.
Agent Workbench is a CLI + Claude Code skill for non-blocking interactions. Claude keeps working while you review and respond in your own time via the web playground.
# Claude runs this (non-blocking)
awb run --title "Deploy" confirm --prompt "Deploy to production?"
# Returns immediately with an ID
# Claude continues working, checks result later
awb wait <id>
claude plugins add-marketplace github:termos-dev/agent-workbench
claude plugins install awb
Then run /awb:init in Claude to configure.
# Launch the playground to monitor and respond
awb ui
# Show help
awb
confirm ask checklist code table markdown html plan-viewer
Additional UI panels can be embedded via html using <awb-*> web components.
MIT
Share bugs, ideas, or general feedback.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Reports Claude Code agent status to OpenUI for real-time status display
Multi-agent tmux dashboard with skills, hooks, and workflow automation
Claude Code の SubAgent(agents/*.md)を正しく定義するためのベストプラクティスガイド。 YAML frontmatter、ツール選択、3-Phase 構造、コンテキスト受け渡し、アンチパターンを網羅。 Use when: agents/*.md を書く、SubAgent 定義を改善する、エージェントの動作が想定外、 コンテキストが渡らない、ツール選択に迷う。 Triggers: "subagent", "agent definition", "agents/*.md", "エージェント定義", "サブエージェント", "3-Phase", "context passing", "コンテキスト渡し", "tool selection", "ツール選択", "subagent_type", "bypassPermissions"
Agent dashboard TUI — tracks AI agent state via hooks for the fleet CLI
Mission Control for Claude Code — auto-starts a web dashboard, provides 90 MCP tools (sessions, stats, live monitoring, projects, teams, insights, coaching, and more), and adds 9 skills including /session-recap, /daily-cost, /standup, /coaching, /insights, and /team-status.
Inter-agent communication for Claude Code and Codex CLI sessions via threads and messages
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claim