Prioritize agent skill invocation against user instructions with precedence rules.
npx skills add https://github.com/koseki2580/skills --skill using-skills-koseki2580
Please help me install this Agent Skill. Skill: using-skills Source: https://github.com/koseki2580/skills/tree/main/skills/using-skills Command: npx skills add https://github.com/koseki2580/skills --skill using-skills-koseki2580
It prevents unnecessary or conflicting skill usage by defining clear rules for when skills should be invoked and how they should be prioritized against user instructions.
Ask the AI to follow the Using Skills guidance: only invoke a relevant skill for the current non-trivial task, and read the full skill instructions before applying any steps.
High-intent search queries and answers about installing and using this skill.
To prioritize agent skills against user instructions, apply a defined precedence order ensuring user-provided instructions override skill behavior. This prevents unnecessary or conflicting skill usage during AI-assisted development and workflow tasks.
Invoke agent skills for task routing only during non-trivial, risky, ambiguous, or specialized work. Do not invoke skills for trivial or obvious edits, applying a minimal-vs-standard decision rule to ensure correctness and safety.
Progressive disclosure for skill invocation requires reading the full skill documentation before executing any steps. This deterministic adoption process ensures the skill fits the task and allows skipping when it turns out to be a poor fit.
Prevent conflicting AI skill usage by defining clear rules for when skills should be invoked and obeying the subagent stop condition. This enforces the precedence order and ensures skill selection impacts correctness, safety, and determinism.
Prompt governance requires reading the full skill documentation before executing it. This deterministic skill adoption process ensures the skill is a good fit and allows skipping when it turns out to be inappropriate for the task.
otto-poblyshInvoke the Skill tool before responding when skill applicability is likely.
FlyFissionEnforce named targets, expected results, and stop conditions for high-stakes agent actions.
TonyQ-AIEnforce a mandatory skill-check protocol before executing AI interactions.
droxeyInvoke the using-superpowers skill before any assistant response or action.
