skills.rest
sek788432
sek788432COMMUNITY
Explore all skills by this author →
GitHub

writing-great-skills

Provides vocabulary and principles for writing predictable AI agent skills.

6|2|Updated May 6, 2026
Search Tags:#documentation#prompt engineering#codex#ai agents#skill authoring#progressive disclosure#predictability
One-click install
npx skills add https://github.com/sek788432/Stock-Back-Test-System --skill writing-great-skills-sek788432
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-great-skills
Source: https://github.com/sek788432/Stock-Back-Test-System/tree/main/.agents/skills/writing-great-skills
Command: npx skills add https://github.com/sek788432/Stock-Back-Test-System --skill writing-great-skills-sek788432

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing skills for AI agents often produces bloated, duplicated, or unreliable instructions that cause inconsistent agent behavior. This Skill supplies a shared vocabulary and a set of principles for structuring, invoking, and pruning skills so the agent follows the same process on every run. ## Core Features & Use Cases - Invocation guidance: Explains model-invoked versus user-invoked skills, how to write trigger-focused descriptions, and when to use a router skill. - Information hierarchy: Defines steps, in-skill reference, and external reference, with progressive disclosure and co-location rules for arranging content. - Failure-mode diagnosis: Catalogs premature completion, duplication, sediment, sprawl, no-ops, and negation, each paired with its cure. - Use Case: When editing a SKILL.md that has grown long and unreliable, use this Skill to identify sprawl and sediment, push reference material behind context pointers, and sharpen completion criteria. ## Quick Start Use the writing-great-skills skill to review my SKILL.md and suggest improvements for predictability and structure.

Frequently Asked Questions about writing-great-skills

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

FAQPage Schema
How do I write a good skill description for an AI agent?▼

Front-load the skill's leading word, list one trigger per distinct branch, and cut identity statements already covered in the body. Every word adds context load, so prune the description even harder than the skill body.

What is the difference between model-invoked and user-invoked skills?▼

A model-invoked skill allows implicit invocation so the agent selects it from its description, while a user-invoked skill is explicit-only and fires only when the human names it. Invocation policy is set in agents/openai.yaml via policy.allow_implicit_invocation.

When should I split a skill into multiple skills?▼

Split by invocation when you have a distinct leading word that should trigger a model-invoked skill on its own. Split by sequence when visible post-completion steps tempt the agent to rush the current step, causing premature completion.

Why does my AI agent skip steps or finish tasks early?▼

This is premature completion, caused by vague completion criteria and visible upcoming steps pulling attention forward. Sharpen the completion criterion first; only hide later steps by splitting if the criterion is irreducibly fuzzy and the rush is observed.

What are the limitations of using negation in skill instructions?▼

Prohibitions name the forbidden behavior into context, making it more available rather than less. Prompt the positive target behavior instead, reserving prohibitions for hard guardrails that cannot be phrased positively.

Related Skills

Explore More Software Engineering Skills→
TyrizxTyrizx

create-instructions

Create .instructions.md files that codify project rules and coding conventions.

Official
Intermediate
tkerseytkersey
66

refine

Apply minimal updates to Codex skill metadata, workflows, and resources.

Community
Advanced
FrogAiFrogAi
1

agent-test-harness

Measure skill and subagent trigger accuracy across multiple prompt trials.

Community
Advanced
luanluan
4

compress-prompt

Compress text content for AI consumption by applying defined rules to reduce token count.

Community
Intermediate