Lobster

Orchestrates multi-step tool pipelines with optional approval checkpoints for repeatable automation.

Updated Apr 7, 2026
One-click install
npx skills add https://github.com/vut08905/openclaw-app --skill lobster-vut08905
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Lobster
Source: https://github.com/vut08905/openclaw-app/tree/main/extensions/lobster
Command: npx skills add https://github.com/vut08905/openclaw-app --skill lobster-vut08905

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Lobster reduces manual coordination by enabling multi-step workflows with built-in approval checkpoints, ensuring repeatable automation and traceable decisions.

Core Features & Use Cases

  • Deterministic, multi-step pipelines with optional approvals before execution.
  • Supports resumable workflows: pause, resume, and continue after user input.
  • Seamless integration with OpenClaw tools and plugin ecosystem to orchestrate complex tasks.

Quick Start

Run a sample workflow with a simple pipeline and an approval gate to see Lobster coordinate steps and return a deterministic result.

Frequently Asked Questions about Lobster

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

FAQPage Schema
How do I build multi-step automation workflows with approval checkpoints?

Multi-step workflows with approval checkpoints are orchestrated by defining tool pipelines with optional human-in-the-loop gates before execution. This ensures repeatable automation and traceable decisions while producing deterministic outcomes.

How do resumable workflows handle pause and resume after user input?

Resumable workflows handle pause and resume by utilizing resumable state management to pause execution, wait for user input, and then continue the pipeline. This ensures safe sandboxed execution and deterministic outcomes after the checkpoint is cleared.

When do I need human-in-the-loop decisions in automated pipelines?

You need human-in-the-loop decisions in automated pipelines for triage, monitoring, and workflows requiring strict oversight across diverse tools. Optional approval checkpoints ensure traceable decisions before executing sensitive or critical pipeline steps.

How do you ensure deterministic outcomes in repeatable automation pipelines?

Deterministic outcomes in repeatable automation pipelines are ensured through strict JSON encapsulation, resumable state management, and safe sandboxed execution with clear error handling. This coordinates multi-step tasks reliably across diverse tools.

Does sandboxed execution with clear error handling support diverse tool ecosystems?

Sandboxed execution with clear error handling supports diverse tool ecosystems by safely isolating multi-step pipeline orchestration. It integrates seamlessly with plugin ecosystems to coordinate complex tasks while maintaining strict JSON encapsulation.