lazy-cloud

Dispatch one AlgoBooth sub-skill per cloud invocation and defer non-cloud steps.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/jacobrocks1212/claude-config --skill lazy-cloud
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lazy-cloud
Source: https://github.com/jacobrocks1212/claude-config/tree/main/repos/algobooth/.claude/skills/lazy-cloud
Command: npx skills add https://github.com/jacobrocks1212/claude-config --skill lazy-cloud

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates cloud-based orchestration of the AlgoBooth '/lazy' workflow by dispatching at most one sub-skill per invocation, while gracefully deferring any step that cannot run in a cloud Linux environment and documenting the deferral for a workstation resumption.

Core Features & Use Cases

  • Cloud-mode wrapper around the /lazy workflow running in ephemeral cloud Linux containers, enforcing the one-skill-per-invocation rule and preserving the same state-machine contracts as the workstation variant.
  • Deferral and sentinel workflow: when a step requires a desktop/MCP environment, the skill writes DEFERRED_NON_CLOUD.md and halts, guiding the workstation to complete testing later.
  • Special actions and operator-free progression: supports actions like write_deferred_non_cloud, write_validated_from_skip, flip_plan_complete_cloud_saturated, and mark_complete to advance or defer parts of the pipeline without multi-step chaining.
  • Status bookends: prints cloud-specific before/after status messages that describe the current feature, the action taken, and what the workstation will do next.

Quick Start

Run the cloud variant for your project to let the state machine dispatch a single sub-skill or issue a deferred action.

Frequently Asked Questions about lazy-cloud

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

FAQPage Schema
How do I defer MCP testing steps in a cloud Linux container to run later on a workstation?

Cloud queue orchestration defers MCP testing steps by writing a DEFERRED_NON_CLOUD.md sentinel file that documents the deferral and halts progression, allowing a workstation to resume testing later.

What is a cloud queue state machine and how does it dispatch one sub-skill per invocation?

Cloud queue orchestration dispatches one sub-skill per invocation by enforcing a strict one-skill-per-invocation rule, preventing multi-step chaining while preserving state-machine contracts in ephemeral Linux containers.

Can I run desktop-dependent MCP workflows in an ephemeral cloud Linux environment?

Desktop-dependent MCP workflows cannot run directly in an ephemeral cloud Linux environment, but operator-free special actions like __write_deferred_non_cloud__ can defer those steps for workstation resumption.

How do I coordinate workstation handoffs after a cloud session defers a pipeline step?

Workstation handoffs after cloud deferral are coordinated through DEFERRED_NON_CLOUD.md sentinel files that document deferred steps, guiding the workstation to complete testing and resume the state machine.

What are the limitations of orchestrating pipeline progression in cloud-only environments?

Orchestrating pipeline progression in cloud-only environments is limited because steps requiring desktop or MCP environments must be deferred, halting the pipeline until a workstation resumes the workflow.