AI Extension · Workflow

Write a Project Skill

Avoid explaining the same process every time. Freeze procedure and validation criteria into a Skill document.

Skill candidates

Start with release prep, PR review, deployment checks, docs updates, or customer-request triage.

Project Skill Drafting Prompt

Copy and use

"I want to turn repeated project work into an AI Skill. Write a SKILL.md draft from the information below. [Repeated work] - Workflow name: [e.g. release prep] - When it runs: [e.g. before deployment] - Inputs: [e.g. branch, environment, change summary] - Files/docs to read: [e.g. package.json, README, docs/deploy.md] - Commands to run: [e.g. npm run build, npm test] - Recovery rule: [e.g. stop and analyze logs if build fails] [Output requirements] 1) Skill purpose 2) Trigger conditions 3) Procedure 4) Validation commands 5) Forbidden actions and approval-required actions 6) Example user requests"

Copy this prompt and paste it into your AI chat.

SKILL.md Skeleton

SKILL.md

# Skill Name

## Purpose

## When to use

## Inputs

## Procedure

## Validation

## Safety

Execution Checklist

0/5 done

Confirm repetitionUse this for workflows repeated three or more times

Write triggerMake when-to-use conditions explicit

Limit read/write scopeList only necessary files and systems

Write validation commandsInclude commands that decide pass/fail

Write approval boundaryDeployment, deletion, and external sends need approval

TL;DR

Use this section as a quick baseline for project skill.
Replace branch names, secrets, URLs, and thresholds with your team-specific values.

Prerequisites

Confirm repository and runtime access permissions.
Verify baseline commands (`install`, `build`, `test`) for local/CI.
Ensure environment variables, secrets, and deployment targets are up to date.

Steps

Lock the change goal and scope in one sentence.
Apply the core project skill changes in minimal commits.
Capture validation logs and immediately update checklist/docs.

Validation

Confirm build/test pass reproducibly on the current branch.
Run manual sample checks on critical user/operation flows.
Cross-check ko/en docs and links for intent consistency.

Troubleshooting

Capture logs, request payloads, and environment values first on failures.
Narrow down by permissions, env vars, routing, and build artifacts.
Record temporary mitigation and root-cause fix separately.

References

AI Skills

Skill vs Tool/MCP

Release/Rollback Runbook

Turn operations into skills

Last updated: June 22, 2026 · Version: v0.0.1

Send Feedback

Opens a new issue page with your message.

Open GitHub Issue