How to use the Blog Writer agent skill (blog-writer)

What the Blog Writer skill makes the agent do

When you ask the agent to write a blog post, this skill makes it work through the same four stages each time. First it reads a bundled style guide. Then it drafts a short outline with a hook, context, 3-5 core sections and a takeaway. Next it writes each section with a clear subheading, short paragraphs and code examples where they help. Finally it polishes transitions, tone and the closing call-to-action.

The skill does not add new software abilities. It is a set of written instructions plus a reference style guide, so the agent uses its existing writing ability under tighter rules.

• Step 1: read references/style-guide.md for style rules
• Step 2: outline with a hook, context, 3-5 core sections and a takeaway
• Step 3: write each section with an H2 subheading, short paragraphs and fenced code blocks with language tags
• Step 4: add transitions, check tone and verify code, then end with a call-to-action or next steps

What the files in the skill contain

The skill is a folder with two files. SKILL.md holds the name, description and the four-step instructions the agent follows. references/style-guide.md holds the writing rules, such as the voice, title and introduction limits, heading levels and formatting conventions.

The style guide is generic. It asks for a conversational but authoritative voice, a title that includes the technology name, an introduction of no more than three paragraphs and complete, runnable code blocks rather than pseudocode.

How to install the Blog Writer skill

You can install the skill for yourself or for one project. Download the ZIP from the skill's page on this site, then unzip it. To make it available to all your projects, unzip it into ~/.claude/skills/. To make it available only inside one project, unzip it into that project's .claude/skills/ folder.

If you use the Claude apps instead, open Settings, go to Capabilities and upload the skill under Skills.

• Personal install: unzip into ~/.claude/skills/
• Project install: unzip into the project's .claude/skills/ folder
• Claude apps: Settings → Capabilities → Skills, then upload the skill

What your agent needs to run it

The skill needs an agent runtime that supports SKILL.md skills and the load_skill_resource tool, because the first step reads the bundled style guide through that tool. If your runtime cannot load skill resources, the agent cannot follow Step 1.

The skill runs no scripts and uses no network access. It only supplies instructions and reference text to the agent.

• An agent runtime that supports SKILL.md skills
• The load_skill_resource tool, to read references/style-guide.md

How to trigger the skill with a natural request

You do not need a special command. The skill is meant to run when you ask for a blog post, so write the request the way you would ask a colleague. Name the topic, the reader and any code you want included. The more specific the request, the better the outline will fit.

Because the skill's steps include a code requirement, mention the technology you want the post to focus on. The style guide asks for a title that names the technology.

• Write a blog post about building a small REST API with Flask, aimed at Python developers, with runnable examples.
• Draft a technical post on caching strategies for a team that already uses Redis. Keep the code in Python.
• Help me write a blog post explaining our new CLI tool for beginners, with an outline first.

Review the output before you publish

The skill asks the agent to verify that code examples are complete and runnable, but the agent can still make mistakes. Run the code yourself, check every command and claim, and confirm that the examples match the versions of the tools your readers use.

The bundled voice is generic. If your publication has a strong or brand-specific voice, plan to edit the draft so it sounds like you.

• Run every code example yourself
• Check facts, commands and links
• Edit the tone to match your own or your brand's voice