intern — for Claude Code intern, Opencode-Intern-Claude-Code-Skill, community, for Claude Code, ide skills, Opencode integration, ACP client automation, task delegation, background execution, synchronous execution, Claude Code

v1.0.0
GitHub

About this Skill

Perfect for AI Agents needing automated task delegation and workflow efficiency using Opencode in Claude through ACP. intern is a Claude Code skill that automates task delegation using Opencode through ACP.

Features

Run ACP client scripts for task delegation
Execute tasks in background mode with immediate task ID return
Check task status and retrieve results upon completion
Wait for task completion with optional timeout
List and filter running tasks for efficient management
Cancel running tasks and clean up completed ones

# Core Topics

smadgerano smadgerano
[1]
[0]
Updated: 1/20/2026

Agent Capability Analysis

The intern skill by smadgerano 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. Optimized for for Claude Code, Opencode integration, ACP client automation.

Ideal Agent Persona

Perfect for AI Agents needing automated task delegation and workflow efficiency using Opencode in Claude through ACP.

Core Value

Empowers agents to automate tasks using Opencode in Claude through ACP, enhancing workflow efficiency and productivity with features like background execution, synchronous execution, and task management using JSON output and protocols like bash.

Capabilities Granted for intern

Automating tasks using Opencode in Claude through ACP
Delegating tasks to run in background or synchronously
Managing tasks using JSON output and checking status

! Prerequisites & Limits

  • Requires Opencode in Claude through ACP
  • Limited to tasks that can be executed using bash
  • Dependent on project_path and prompt for task execution
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

intern

Unlock efficient task delegation with the intern AI agent skill, utilizing Opencode in Claude through ACP for seamless automation and enhanced productivity.

SKILL.md
Readonly

Intern

Run the ACP client script to delegate tasks:

<skill-path> is this skills location on disk.

Background Execution (Default)

Tasks run in background by default, returning a task ID immediately:

bash
1# Launch task (background is default) 2python <skill-path>/scripts/acp_client.py "<project_path>" "<prompt>" --json 3# Returns: {"task_id": "abc123", "status": "launched"} 4 5# Check status 6python <skill-path>/scripts/acp_client.py --status abc123 --json 7 8# Get result when complete 9python <skill-path>/scripts/acp_client.py --result abc123 --json 10 11# Wait for completion (blocking) 12python <skill-path>/scripts/acp_client.py --wait abc123 --json --timeout 600 13 14# List all tasks 15python <skill-path>/scripts/acp_client.py --list-tasks --json 16python <skill-path>/scripts/acp_client.py --list-tasks --filter running --json 17 18# Cancel a running task 19python <skill-path>/scripts/acp_client.py --cancel abc123 20 21# Clean up old completed tasks 22python <skill-path>/scripts/acp_client.py --cleanup

Synchronous Execution

For simple tasks where you want to wait for the result:

bash
1python <skill-path>/scripts/acp_client.py "<project_path>" "<prompt>" --sync --json

Parse the JSON result directly. The project_path is usually the current working directory.

Response Format

json
1{ 2 "success": true, 3 "response": "Agent's response text", 4 "tools": [{"name": "bash", "status": "completed"}], 5 "errors": [], 6 "agent_status": "end_turn" 7}

Agent Status Values

StatusMeaning
end_turnNormal successful completion
idleAgent finished and is idle
refusalAgent refused to complete the task
cancelledSession was cancelled
max_tokensResponse truncated due to token limit
max_turn_requestsHit maximum turn limit
errorAgent encountered an error
unknownStatus not determined

Task Status Values (Background Mode)

StatusMeaning
pendingTask created, not yet started
runningTask is executing
completedTask finished successfully
failedTask failed with errors
cancelledTask was cancelled by user

Parameters

Execution Options

  • project_path (required for sync): Absolute path to project
  • prompt (required for sync): Task description (max 100KB)
  • --agent: Agent name (default: intern)
  • --timeout: Overall timeout in seconds (default: 300)
  • --activity-timeout: Max seconds without activity (default: 120)
  • --json: JSON output (always use this)
  • --debug: Enable debug logging to stderr
  • --background: Run task in background (default)
  • --sync: Run task synchronously, wait for completion

Task Management Options

  • --status TASK_ID: Check status of a background task
  • --list-tasks: List all background tasks
  • --filter STATUS: Filter tasks by status (pending/running/completed/failed/cancelled)
  • --result TASK_ID: Get the result of a completed task
  • --cancel TASK_ID: Cancel a running task
  • --wait TASK_ID: Wait for a task to complete
  • --cleanup: Remove old completed tasks (>24 hours)

Errors

ErrorFix
"OpenCode not found in PATH"Ensure opencode is installed and in PATH
"Overall timeout"Increase --timeout or simplify task
"Activity timeout"Agent may be hung; try simplifying task
"Agent process died"Check opencode installation and logs
"Agent error"Service issue; check opencode auth and API status
"Prompt exceeds maximum length"Prompt is >100KB; break into smaller tasks
"Agent refused to complete"Task may violate agent policies; rephrase

Features

  • Thread-safe request handling: Safe for concurrent operations
  • Background task execution: Launch tasks asynchronously, check status later
  • Activity timeout detection: Detects hung agents that stop responding
  • Process health monitoring: Detects if agent process crashes
  • Session cancellation: Properly cancels stuck sessions per ACP protocol
  • Error status detection: Detects and reports agent errors
  • Debug mode: Enable --debug for troubleshooting
  • Partial response capture: Returns partial responses even on failure
  • Windows path normalization: Automatically converts C:\path to C:/path in prompts

Best Practices

  1. Always use --json for reliable parsing of results
  2. Use background mode for tasks >2 minutes to avoid blocking
  3. Check task status periodically rather than using long timeouts
  4. Use --debug when troubleshooting to see protocol messages
  5. Keep prompts focused - delegate one clear task at a time
  6. Clean up old tasks periodically with --cleanup

FAQ & Installation Steps

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

? Frequently Asked Questions

What is intern?

Perfect for AI Agents needing automated task delegation and workflow efficiency using Opencode in Claude through ACP. intern is a Claude Code skill that automates task delegation using Opencode through ACP.

How do I install intern?

Run the command: npx killer-skills add smadgerano/Opencode-Intern-Claude-Code-Skill. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for intern?

Key use cases include: Automating tasks using Opencode in Claude through ACP, Delegating tasks to run in background or synchronously, Managing tasks using JSON output and checking status.

Which IDEs are compatible with intern?

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 intern?

Requires Opencode in Claude through ACP. Limited to tasks that can be executed using bash. Dependent on project_path and prompt for task execution.

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 smadgerano/Opencode-Intern-Claude-Code-Skill. 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 intern immediately in the current project.

Related Skills

Looking for an alternative to intern 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