ask-claude

Capture Claude CLI outputs as structured, reusable Markdown artifacts via Forge CLI.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/lpjhelder/kronon-forge --skill ask-claude-lpjhelder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ask-claude
Source: https://github.com/lpjhelder/kronon-forge/tree/main/skills/ask-claude
Command: npx skills add https://github.com/lpjhelder/kronon-forge --skill ask-claude-lpjhelder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Running a local Claude CLI and manually capturing its responses can be error-prone and inconsistent; this Skill standardizes execution and preserves outputs as searchable, reusable artifacts so insights and action items are not lost after an interactive session.

Core Features & Use Cases

  • Local CLI Execution: Invoke the user's locally installed Claude binary via the Forge CLI to get direct, deterministic responses without MCP routing.
  • Structured Artifact Generation: Save each interaction as a markdown artifact that includes the original task, the final prompt, raw Claude output, a concise summary, and actionable next steps.
  • Graceful Missing-Binary Handling: Detect when the Claude binary is absent, explain installation requirements, and provide a verification command to the user.
  • Use Case: Ask Claude for a focused code review or second opinion, capture the results for team review, and track follow-up action items in project state.

Quick Start

Ask Claude locally to review the authentication change and produce a concise summary with action items.

Frequently Asked Questions about ask-claude

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

FAQPage Schema
How do I save Claude CLI responses as reusable markdown artifacts?

You can save Claude CLI responses as reusable markdown artifacts by executing the query through a local CLI wrapper that captures the original task, final prompt, raw output, summary, and action items into a structured file.

Do I need a local Claude binary installed to query Claude without remote routing?

Yes, you need a functioning local Claude binary installed to query Claude without remote routing. If the binary is missing, the execution will gracefully fail and provide the specific installation requirements and a verification command.

What is the best way to capture a local Claude code review for team project state?

The best way to capture a local Claude code review for team project state is to run the query via the Forge CLI, which standardizes execution and preserves the output, concise summary, and action items as a searchable markdown artifact.

How does structured artifact generation work for developer workflow prompting?

Structured artifact generation for developer workflow prompting works by executing a focused prompt locally and saving the interaction as a markdown file containing the original task, final prompt, raw output, summary, and actionable next steps.

What happens if the Claude binary is not found during local CLI execution?

If the Claude binary is not found during local CLI execution, the process detects the absence, explains the installation requirements to the user, and provides a specific verification command to ensure the environment is properly configured.

Can I get a second opinion on code changes using a local Claude CLI?

Yes, you can get a second opinion on code changes using a local Claude CLI by applying this workflow to execute focused questions, capturing the deterministic response alongside action items in a reusable markdown artifact for future reference.