llm-review

Orchestrates local LLM task lifecycle including load, execute script, collect results, unload via agent dispatch.

17|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/alexiolan/craft-skills --skill llm-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: llm-review
Source: https://github.com/alexiolan/craft-skills/tree/main/skills/llm-review
Command: npx skills add https://github.com/alexiolan/craft-skills --skill llm-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

End-to-end orchestration of local LLM tasks.

Core Features & Use Cases

  • Lifecycle management of local LLM tasks: availability check, model loading, script execution, result collection, and model unloading.
  • Agent-based dispatch: instructs other skills to dispatch as an agent rather than running LLM bash commands directly.
  • Safe, observable execution: clear steps and accountability for each phase of the workflow.

Quick Start

Invoke a local LLM review on a target file to analyze its security focus.

Frequently Asked Questions about llm-review

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

FAQPage Schema
How do I orchestrate a local LLM task lifecycle from loading to unloading?

Local LLM task lifecycle orchestration is handled by enforcing an availability check, model loading, script execution, result capture, and clean model unloading using a dedicated agent.

What is the best way to dispatch local LLM scripts without running inline bash commands?

Dispatching local LLM scripts without inline bash is achieved by using an agent-based approach that instructs other skills to execute tasks, ensuring safe and observable workflow phases.

Can I run local LLM model reviews in the background and explicitly unload the model afterwards?

Running local LLM reviews in the background with explicit unload controls is supported, allowing you to manage model lifecycle phases and release resources safely after result collection.

Does local LLM lifecycle management require an availability check before script execution?

Local LLM lifecycle management requires an availability check before script execution to verify the model is ready, ensuring safe task dispatch and preventing execution errors.

Why use an agent for local LLM task dispatch instead of direct bash execution?

Using an agent for local LLM task dispatch provides clear accountability and observable execution steps across the lifecycle, avoiding the risks of running direct bash commands inline.

How do I perform a security review on a target file using a local LLM?

Performing a security review on a target file using a local LLM involves invoking the agent to load the model, execute the analysis script, capture results, and cleanly unload.