Skip to content

Debanitrkl/technical-writing-automation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Technical Writing Automation

A plug-and-play toolkit of agent skills and workflows that you can drop into any repo and use with Claude Code or OpenAI Codex CLI to plan, draft, and review technical writing at the same quality bar we use at Parseable.

No coding required. If you can copy a folder and paste it into another folder, you can use this.


What you get

Folder What it is Why you'd use it
.agents/skills/ 4 ready-to-use writing skills Turn your AI coding assistant into a careful editor
.claude/workflows/ Step-by-step blog creation playbook Never miss frontmatter, images, or SEO fields

The 4 skills

Each skill lives in .agents/skills/<skill-name>/SKILL.md. An AI assistant reads the SKILL.md and applies its rules to your writing.

  • review-writing — structured editorial review. Catches weak claims, unclear arguments, and missing evidence.
  • seo-aeo-best-practices — modern SEO + Answer Engine Optimization (AEO). Makes content rankable and quotable by AI answer engines.
  • stop-slop — strips the telltale "AI-sounding" filler: hedges, throat-clearing intros, "In today's fast-paced world…" openers.
  • writing-clearly-and-concisely — Strunk & White-style concision pass. Cuts flab, strengthens verbs, tightens sentences.

Who this is for

  • Founders, marketers, DevRel, solo writers who want their AI assistant to stop writing like a bot.
  • Engineering teams publishing technical content and tired of reviewing the same AI-generated clichés.
  • Anyone who has Claude Code or Codex installed and a folder of writing to improve.

You do not need to know JavaScript, Python, or how to use git beyond git clone (and we cover that below).


Two ways to use this

Option A — Fork the repo (recommended if you want updates)

  1. Click Fork at the top-right of https://github.com/Debanitrkl/technical-writing-automation
  2. Clone your fork to your computer:
    git clone https://github.com/<your-username>/technical-writing-automation.git
    cd technical-writing-automation
  3. Open the folder in Claude Code or Codex and start working (see "Using the skills" below).

Option B — Copy-paste into your existing repo (fastest)

  1. Download this repo as a ZIP: https://github.com/Debanitrkl/technical-writing-automation/archive/refs/heads/main.zip
  2. Unzip it.
  3. Copy these two folders into the root of your project:
    • .agents/ → gives you the skills
    • .claude/ → gives you the workflow playbook
  4. That's it. Your repo now has the skills. Commit them if you want them shared with teammates:
    git add .agents .claude
    git commit -m "Add technical writing skills"

The leading dot (.) in .agents and .claude is intentional — these are "hidden" config folders that AI tools look for automatically. On macOS Finder, press Cmd+Shift+. to see hidden folders.


Using the skills

With Claude Code

  1. Install Claude Code (one-time): https://claude.com/claude-code
  2. Open your project in a terminal and run:
    claude
  3. Ask Claude to use a skill in plain English. Examples:
    • review this draft using the review-writing skill: drafts/my-post.md
    • run the stop-slop skill on the intro I just wrote
    • apply seo-aeo-best-practices to blog/launch.md
    • use writing-clearly-and-concisely on the README
  4. Claude will read the relevant SKILL.md, apply its rules, and either suggest edits or rewrite inline. You review, accept, or push back.

With OpenAI Codex CLI

  1. Install Codex CLI (one-time): https://github.com/openai/codex
  2. Run it in your project folder:
    codex
  3. Ask Codex to read a skill and apply it. Codex doesn't auto-load .agents/skills/ the way Claude Code does, so point it at the file:
    Read .agents/skills/stop-slop/SKILL.md and apply it to drafts/post.md
    
    Follow .agents/skills/seo-aeo-best-practices/SKILL.md when writing
    the outline for my new post about log management.
    
  4. Codex will follow the instructions inside the SKILL.md file and edit accordingly.

With any other AI assistant (Cursor, Continue, Aider, ChatGPT web, etc.)

The skills are just markdown files. You can paste the contents of a SKILL.md directly into any chat:

Here are the rules I want you to follow when reviewing my writing:

<paste the full contents of .agents/skills/review-writing/SKILL.md>

Now review this draft:

<paste your draft>

The blog workflow

.claude/workflows/create-blog-post.md is a checklist for creating a blog post — originally written for the Parseable website, but the structure (frontmatter fields, author registry, image specs, local preview, commit) maps to any Next.js/Hugo/Jekyll/Astro blog.

How to use it:

  • Claude Code: follow the create-blog-post workflow for a post titled "…"
  • Codex: read .claude/workflows/create-blog-post.md and walk me through it
  • Manually: open the file and follow the steps.

Adapt the frontmatter fields and author list in the workflow to match your own site.


FAQ

Do I need a paid Claude or OpenAI account? Yes — both Claude Code and Codex require API access (or a subscription). See their setup pages linked above.

Can I change the skills? Yes. They're plain markdown — open any SKILL.md, edit the rules, save. The AI will pick up your edits next time you invoke the skill.

Can I add new skills? Yes. Create .agents/skills/<your-skill-name>/SKILL.md and describe when and how it should be applied. Both Claude Code and Codex can read it.

Is this affiliated with Anthropic or OpenAI? No. This repo bundles independently-written skills that happen to work well with their tools.


License

Individual skills may carry their own LICENSE file (e.g. stop-slop/ bundles one). The rest of this repo is provided as-is for public use — fork it, copy it, adapt it.


Questions / issues

Open an issue on this repo, or reach out via the Parseable team.

About

Agent skills, workflows, and SEO content playbook for Parseable technical writing.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors