| name | ai-skill-writer |
| description | Turn recurring tasks, SOPs, workflows, prompts, business processes, or expert know-how into reusable AI skills with trigger descriptions, step-by-step instructions, constraints, examples, and quality checks. Use when a user wants to create a Codex skill, Claude skill, agent skill, prompt pack, or reusable AI workflow. |
AI Skill Writer
Convert recurring work into a reusable skill an AI agent can reliably invoke and execute.
Workflow
- Identify the recurring job: what the user asks for, what artifact they provide, and what output they expect.
- Write concrete trigger language. Include task names, source artifact types, and situations where the skill should be used.
- Extract the expert workflow:
- intake
- analysis
- transformation
- output structure
- validation
- edge cases
- Decide what belongs in the skill body versus external references:
- body: essential workflow and output rules
- references: long examples, schemas, policies, brand guides
- scripts: deterministic repeated code
- assets: templates used in final outputs
- Write concise instructions. Assume the agent is smart; include only the non-obvious process knowledge.
- Add quality checks and failure handling.
Output Format
# Skill Design: <skill name>
## Skill Name
<lowercase-hyphen-name>
## Trigger Description
<frontmatter-ready description>
## Core Workflow
1. <step>
## Output Contract
<required sections or artifact format>
## Edge Cases
- <case> -> <handling>
## Quality Checks
- <check>
## Draft SKILL.md
\`\`\`markdown
---
name: <skill-name>
description: <description>
---
# <Title>
<instructions>
\`\`\`
Quality Bar
- Make trigger descriptions comprehensive because they decide whether the skill loads.
- Keep skill bodies concise and procedural.
- Do not create giant prompt essays.
- Include examples only when they materially improve execution.