awb — community agent-workbench, community, ide skills, Claude Code, Cursor, Windsurf

v1.0.0
GitHub

About this Skill

Perfect for Claude Code Agents needing interactive development and debugging capabilities with JSON response handling. Interactive Development Environment for Claude Code

termos-dev termos-dev
[0]
[0]
Updated: 3/5/2026

Agent Capability Analysis

The awb skill by termos-dev is an open-source community AI agent skill for Claude Code and other IDE workflows, helping agents execute tasks with better context, repeatability, and domain-specific guidance.

Ideal Agent Persona

Perfect for Claude Code Agents needing interactive development and debugging capabilities with JSON response handling.

Core Value

Empowers agents to run interactive components like confirm, checklist, and ask, while displaying content with code, table, markdown, and HTML, utilizing the awb run command with --title specification and awaiting JSON responses with awb wait.

↓ Capabilities Granted for awb

Running interactive confirm components for user input validation
Displaying code snippets with syntax highlighting using the code component
Debugging agent workflows by awaiting JSON responses with awb wait

! Prerequisites & Limits

  • Requires the awb ui to be open
  • Needs --title specification for every awb run command
  • Limited to components provided by the Agent Workbench (awb)
Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

awb

Install awb, an AI agent skill for AI agent workflows and automation. Works with Claude Code, Cursor, and Windsurf with one-command setup.

SKILL.md
Readonly

Agent Workbench (awb)

Quick Start

  1. Ensure the user has the playground open with awb ui.
  2. Run a component with awb run --title "..." <component> [args].
  3. Always call awb wait <id> and read the JSON response.

Components

Interactive: confirm, checklist, ask Display: code, table, markdown, plan-viewer, html

Essential Rules

  • Require --title for every awb run.
  • Run awb run --help before using a component you are unsure about.
  • If awb is not on PATH, ask the user to install or run the local CLI.

Interactive Runs (Always wait)

awb run returns immediately with an ID. Always call awb wait <id> to get the response.

bash
1ID=$(awb run --title "Q1" confirm --prompt "Approve?")
2awb wait $ID

Fire multiple at once, then wait for each:

bash
1ID1=$(awb run --title "Q1" confirm --prompt "Approve?")
2ID2=$(awb run --title "Q2" checklist --items "A,B,C")
3awb wait $ID1
4awb wait $ID2

Wait Results

awb wait <id> returns JSON with the user's response:

json
1{"confirmed": true}
2{"items": [{"text":"A","checked":true}]}
3{"answers": {"key": "value"}}

Examples

confirm

bash
1awb run --title "Deploy" confirm --prompt "Deploy to production?"

checklist

bash
1awb run --title "Tasks" checklist --items "Build,Test,Deploy"

ask (1-4 questions)

bash
1awb run --title "Name" ask --questions '[{"question":"Your name?","header":"name"}]'
2awb run --title "Pick" ask --questions '[{"question":"Language?","header":"lang","options":[{"label":"TypeScript"},{"label":"Python"}]}]'
3awb run --title "Features" ask --questions '[{"question":"Enable?","header":"feat","multiSelect":true,"options":[{"label":"Auth"},{"label":"API"}]}]'

table

bash
1awb run --title "Data" table --data '[{"name":"Alice","age":30},{"name":"Bob","age":25}]'

code (file only)

bash
1awb run --title "Source" code --file ./src/index.ts

markdown

bash
1awb run --title "Docs" markdown --file ./README.md
2awb run --title "Notes" markdown --content "# Hello\n\nWorld"

html

bash
1awb run --title "Config" html --content '<html>...</html>'

plan-viewer

bash
1awb run --title "Plan" plan-viewer --file ./plan.md

Background Processes with tmux

Use tmux for long-running processes. Run awb --help to see your tmux session name.

bash
1tmux new-session -A -d -s <session-name>
2tmux new-window -t <session-name> -n "dev" "npm run dev"
3tmux capture-pane -t <session-name>:dev -p
4tmux send-keys -t <session-name>:dev "npm test" Enter
5tmux kill-window -t <session-name>:dev

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is awb?

Perfect for Claude Code Agents needing interactive development and debugging capabilities with JSON response handling. Interactive Development Environment for Claude Code

How do I install awb?

Run the command: npx killer-skills add termos-dev/agent-workbench. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for awb?

Key use cases include: Running interactive confirm components for user input validation, Displaying code snippets with syntax highlighting using the code component, Debugging agent workflows by awaiting JSON responses with awb wait.

Which IDEs are compatible with awb?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for awb?

Requires the awb ui to be open. Needs --title specification for every awb run command. Limited to components provided by the Agent Workbench (awb).

↓ How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add termos-dev/agent-workbench. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use awb immediately in the current project.

Related Skills

Looking for an alternative to awb or another community skill for your workflow? Explore these related open-source skills.

View All

widget-generator

Logo of f
f

f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

★ 149.6k
⑂ 0
AI

flags

Logo of vercel
vercel

flags is a Next.js feature management skill that enables developers to efficiently add or modify framework feature flags, streamlining React application development.

★ 138.4k
⑂ 0
Browser

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

★ 72.8k
⑂ 0
AI

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

★ 72.8k
⑂ 0
AI