plan-tune

Tune per-question prompt sensitivity across gstack skills with never-ask, always-ask, or ask-only-for-one-way preferences.

1|Updated May 27, 2026

Search Tags:#interactive #prompts #AskUserQuestion #developer-experience #gstack #question-tuning #plan-tune

Ready to run

npx skills add https://github.com/mi179/claude-skills --skill plan-tune-mi179

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: plan-tune
Source: https://github.com/mi179/claude-skills/tree/main/user/gstack/gstack/plan-tune
Command: npx skills add https://github.com/mi179/claude-skills --skill plan-tune-mi179

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reduces the friction of tuning prompts by allowing per-question control across gstack skills, enabling smarter, lighter interactions for developers.

Core Features & Use Cases

Review which AskUserQuestion prompts fire across gstack skills and selectively enable or disable them per question.
Set per-question preferences: never-ask / always-ask / ask-only-for-one-way to tailor the dialogue.
Inspect and align the dual-track profile (declared user goals vs observed behavior) and adjust tuning accordingly.
Maintain plan-mode safety and respect session-state constraints while applying tuning changes.

Quick Start

Ask Claude to run plan-tune and specify your per-question preferences to tailor gstack prompts.

Frequently Asked Questions about plan-tune

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

FAQPage Schema

How do I control interactive prompts so they stop asking unnecessary questions during development?▼

You can set per-question preferences such as never-ask, always-ask, or ask-only-for-one-way to control exactly which interactive prompts fire or stay silent across gstack skills, reducing dialogue friction during development.

Can I disable specific AskUserQuestion prompts without turning off all proactive prompts?▼

Yes, you can selectively disable individual AskUserQuestion prompts per question while still respecting gating rules like PROACTIVE settings and safety checks, keeping broader proactive prompts active.

How do I align prompt tuning with my developer profile and plan-mode constraints?▼

Prompt tuning operates within the dual-track developer profile (declared goals vs observed behavior) and strictly honors plan-mode constraints and session state to safely apply and persist per-question preference changes.

What is the best way to configure gstack question sensitivity without using CLI syntax?▼

Simply ask Claude to run plan-tune and specify your per-question preferences; the system operates without requiring CLI syntax, persisting changes automatically while honoring spawned session gating and safety checks.

Does prompt tuning work safely within spawned sessions and plan-mode?▼

Prompt tuning respects plan-mode safety constraints, session-state rules, and gating mechanisms like spawned sessions and PROACTIVE checks, ensuring preference changes persist safely without breaking active development workflows.