skills.rest
EveryInc
EveryIncOFFICIAL
Explore all skills by this author →
WebsiteGitHub

ce-skill-work

Applies a skill-authoring standard to create, edit, review, and respond to feedback on SKILL.md files.

24.7k|2.0k|Updated Oct 9, 2025
Search Tags:#skill.md#documentation standards#Prompt Engineering#Code review#Claude code#AI Agents#SKILL authoring
One-click install
npx skills add https://github.com/EveryInc/compound-engineering-plugin --skill ce-skill-work
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ce-skill-work
Source: https://github.com/EveryInc/compound-engineering-plugin/tree/main/.agents/skills/ce-skill-work
Command: npx skills add https://github.com/EveryInc/compound-engineering-plugin --skill ce-skill-work

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Maintaining AI agent skills consistently is hard: SKILL.md files drift into case enumerations, duplicated commands, and misplaced mechanisms. This Skill applies one authoring standard as a procedure whenever you create, edit, review, or respond to review feedback on any file under skills/**.

Core Features & Use Cases

  • Four operating modes: creating a new skill, editing an existing one, reviewing a skill change, and acting on review feedback, each with its own reference guide and done condition.
  • Standards enforcement: rules for conditions-over-cases, provenance-based removal, layered mechanism ownership, portable descriptions, and cross-model (Claude/Codex) compatibility.
  • Validation guidance: mechanical contracts go to bun test while behavior-bearing prose changes get targeted evals on multiple hosts.
  • Use Case: When a bot review flags a skill file, use this Skill to classify each finding as Change/Verify/Consider, close gaps at the owning layer, and record the verdicts.

Quick Start

Use the ce-skill-work skill to review the changes I made to the SKILL.md file in my skills directory.

Frequently Asked Questions about ce-skill-work

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

FAQPage Schema
How do I create a new AI agent skill with a SKILL.md file?▼

Write the outcome spine first (result, consumer, done condition), then the activation contract with a context-pointer description, then layer protocol only where omission causes wrong paths. Update repo inventory docs and run activation, restraint, and main-path evals on Claude and Codex.

How do I review changes to a SKILL.md file?▼

Classify every finding as Change, Verify, or Consider with the evidence each class requires. State requested fixes as conditions or owning-layer moves, never as cases to add, and check that a case is not already decided by a stated condition before filing it.

When should skill instructions be edited versus left alone?▼

Edit when a condition is wrong or missing, or a mechanism sits at the wrong layer. Search provenance first: lines backed by tests, documented learnings, or bug-fix commits stay; lines without provenance are tested against admission criteria before removal.

Does this skill apply to regular source code changes?▼

No. It explicitly covers only files under skills/** such as SKILL.md, references, persona prompts, and bundled script instructions. Changes to src/, tests/, or scripts/ code receive ordinary code review instead.

How are skill behavior changes validated?▼

Mechanical contracts like frontmatter and paths go in bun test. Behavior-bearing prose changes get a targeted eval injecting the on-disk skill into a fresh agent via the host CLI, run on Claude and Codex, with skip reasons recorded when a capability is unavailable.

Related Skills

Explore More Software Engineering Skills→
sartoris-digitalsartoris-digital

using-superpowers

Guide AI to discover, invoke, and prioritize specialized skills in conversations.

Official
Basic
telagodtelagod
238

verify-change

Analyze workspace changes to detect documentation drift and missing tests.

Community
Intermediate
misty-stepmisty-step
2

harness

Coordinate and enforce cross-harness discipline for agent skills and infrastructure.

Official
Advanced
XAIHTXAIHT
14

skill-creator

Bootstrap Tlamatini skill packages with SKILL.md scaffolding and linting.

Official
Intermediate