openhands

Automate coding tasks via OpenHands CLI with LiteLLM providers.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/THTProtocol/lastochka --skill openhands-thtprotocol
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openhands
Source: https://github.com/THTProtocol/lastochka/tree/main/optional-skills/autonomous-ai-agents/openhands
Command: npx skills add https://github.com/THTProtocol/lastochka --skill openhands-thtprotocol

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python, openhands-ai, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill simplifies coding tasks by leveraging the OpenHands CLI, which supports model-agnostic LiteLLM providers, reducing the complexity of coding operations.

Core Features & Use Cases

  • Model-Agnostic Coding: Utilizes any LiteLLM-supported provider (OpenAI, Anthropic, OpenRouter, DeepSeek, Ollama, vLLM, etc.) for coding tasks.
  • Headless Mode: Executes batch/one-shot delegation without interactive UI, ideal for automation.
  • Use Case: When a user needs to delegate a coding task to OpenHands and prefers a coding agent that can run on non-Anthropic/Non-OpenAI providers, this skill is suitable for multi-step file edits and shell commands within a workspace.

Quick Start

Delegate a coding task to OpenHands using the terminal tool with the command: terminal(command="openhands --headless --json --override-with-envs --exit-without-confirmation -t 'Add error handling to all API calls in src/'").

Frequently Asked Questions about openhands

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate coding tasks using a headless agent?▼

Automate coding tasks by delegating multi-step file edits and shell commands to a headless agent. This Skill executes batch operations without an interactive UI, enabling automated workspace modifications.

Can I use OpenHands with non-OpenAI or Anthropic model providers?▼

Yes, OpenHands supports model-agnostic LiteLLM providers. You can use OpenRouter, DeepSeek, Ollama, vLLM, or any other LiteLLM-supported vendor by configuring specific environment variables for the model and API key.

What is the best way to run one-shot code delegation for batch processing?▼

Run one-shot code delegation by executing the OpenHands CLI in headless mode with JSON output. Passing the task string and overriding environment variables allows immediate execution without interactive confirmation.

Do I need Python to run OpenHands for LiteLLM coding automation?▼

Yes, Python 3.12 or higher is required to run OpenHands coding automation. You must also install the openhands-ai dependency and configure your LiteLLM provider environment variables.

How does model-agnostic coding automation handle file edits and shell commands?▼

Model-agnostic coding automation executes multi-step file edits and shell commands within a workspace by sending instructions to the configured LiteLLM provider. The headless CLI processes the task and exits without confirmation.

What are the limitations of using headless mode for automated coding tasks?▼

Headless mode is limited to non-interactive batch execution, meaning it cannot handle real-time user prompts during execution. It requires precise task instructions and pre-configured environment variables for the LiteLLM provider.