blog-post-writing
Write a blog post that reads like a person wrote it and earns search traffic.
Skill by Tam Nguyen
Chalk Labs
Need help deploying blog-post-writing?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/amahony-blog-post-writing
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing, marketing, sales
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/blog-post-writing
What's inside
- Skill name
blog-post-writing- Description
- Write a blog post that reads like a person wrote it and earns search traffic. Use when someone wants to draft, edit or plan a blog post, write about their product or project for search and traction, or make an existing draft sound less like AI. Picks an angle that no published post already owns, builds a lead and section spine, drafts in a plain house voice, runs a humanizer pass, and adds on-page SEO and a distribution plan.
- License
- MIT
- Compatibility
- Uses library_search and library_read for published posts, brand guidelines and product facts, and library_write to save drafts. Web search, when the model has it, helps check what already ranks.
- Metadata
- author: Tam Nguyen · source: https://github.com/tamdogood/builder-essential-skills/tree/main/skills/write-blog · version: 1.0.0 · category: Writing
- Instructions
- 71 lines in SKILL.md · view on GitHub
Files (6)
- LICENSE
- SKILL.md
- angles.md · references/
- craft.md · references/
- seo.md · references/
- voice.md · references/
SKILL.md (first lines)
# Blog post writing
The bar: the post reads like a person wrote it, it targets a search cluster no other post of theirs already owns, and every claim in it is true.
Read the references with skill_read as each step calls for them: `references/angles.md`, `references/craft.md`, `references/voice.md` and `references/seo.md`. `voice.md` is the default house voice. When the organisation has a style guide in the library, it wins wherever the two disagree.
## 1. Read the ground truth first
Before writing, look in the library with library_search and library_read for:
- **Published posts:** every title, description and tag set, so the new post does not compete with one. If the library does not hold them, ask for the blog's URL or a list.
- **One or two full posts**, to absorb the voice.
- **Product facts, docs and brand guidelines**, so claims and examples are accurate.
Ask only for what is missing.
## 2. Pick and pitch the angle
Each post must own a distinct search cluster; two posts chasing the same keywords split their own ranking. Using `angles.md`, pick an angle no published post owns, then pitch it in one message before drafting: the working title, the one-sentence thesis and the three to five keyword phrases it targets. Wait for a yes. If the topic collides with an existing post, say so and propose an adjacent, distinct angle.
## 3. Outline
A good post has a spine, not a listicle. Read `craft.md` before outlining; the lead and the heading spine decide most of whether the post is finished and found.
- **A lead** (description and opening) that names a concrete, specific problem the reader has.
- **Four to seven sections**, each making a claim backed by a real example, command, number or quote from the person's material. Made-up product details are worse than none.
- **An honest-limitations beat**: what this is not, or what is deferred. Candour keeps it from reading like marketing.
- **A close** that lands the thesis and ends on one concrete thing the reader can do or check, never a generic wrap-up.
Read only the headings in order: they must show the argument.
## 4. Draft
Write plain markdown. Follow `voice.md` and `craft.md`. Keep one clear opinion and support it with evidence. For technical or instructional passages, apply the Simplified Technical English baseline in `voice.md`: short sentences, one term per concept, direct actions, exact code and names. Apply the direct-response pass from `craft.md` to the title, description and closing call to action only.
## 5. Humanize (mandatory)
Run the humanizer checklist in `voice.md` over the whole draft. Strip significance inflation, promotional adjectives, the rule of three, trailing "-ing" analyses, negative parallelisms ("it's not X, it's Y"), synonym cycling, hedging and chat artefacts. Check the hard rules last, including the search for dashes and curly quotes, then read it as if aloud and fix anything hard to follow.
## 6. Add the SEO and hand over
Follow `seo.md`:
- Put the primary phrase in the title, the description, the first heading and the tags.
- Suggest two or three internal links to related published posts with descriptive anchor text, one reciprocal link from the most related existing post, and one outbound link to an authority.
- Give the post's metadata in one block: title, slug, description, tags and a cover image idea.
You cannot publish the post or open the site. Give the person the post-publish checklist from `seo.md` (sitemap, feed, meta tags, structured data, cover image) to run themselves.
## 7. Distribution plan
A correct post with no inbound links does not rank. After the draft, offer:
- a short social thread that teaches the core insight and links the post's production URL;
- where the question the post answers is already being asked, so the person can answer it there and link the post;
- which docs or pages should link to it.
Offer to write the thread, or hand it to the matching skill, such as linkedin-post-writing for LinkedIn.
## Deliver
Return the pitch first and wait for agreement. After that, return the metadata block, the post in markdown, the SEO suggestions, the checklist and the distribution plan, in that order. Mark placeholders for details only the author knows. Save the draft with library_write only when someone asks, and say the file name.
## Guardrails
- Never invent product features, figures, quotes, customers or results. Quote the person's material.
- Never weaken the voice to sound "professional". Concrete and a little contrarian beats polished and generic.
- Do not write a post whose angle overlaps a published one; adjust the angle.
- Do not claim strict Simplified Technical English conformance.From amahony/agent-templates, MIT.
More from amahony/agent-templates
- 0.570score
paper-opportunity-radar
Tam Nguyen
Run a cumulative daily or one-off sweep of research papers on a chosen topic, audit their claims, methods, integrity signals and independent support, then…
SkillAFreshness: AClaude Code +9 - 0.570score
business-naming
Tam Nguyen
Generate, refine, compare and, when needed, screen distinctive names for startups, products, apps, developer tools, open-source projects, product families…
SkillAFreshness: AClaude Code +9 - 0.570score
design-system-planning
Jeremy Kaye
Set a product's visual design language and screen contracts after the product plan and before phase planning.
SkillAFreshness: AClaude Code +9 - 0.570score
mastery-plan
Tam Nguyen
Build an evidence-based path toward exceptional capability in a field or skill, and coach the learner along it.
SkillAFreshness: AClaude Code +9 - 0.570score
phase-planning
Jeremy Kaye
Create and maintain phase files, the plans that break work into phases, sprints and tasks with a status for every task.
SkillAFreshness: AClaude Code +9 - 0.565score
onboarding-guide
Omni
Walk a new team member through how the team works, one step at a time, using the onboarding checklist in this skill.
SkillAFreshness: AClaude Code +9 - 0.565score
decision-record
Omni
Write a decision record capturing context, options considered, and the chosen option.
SkillAFreshness: AClaude Code +9 - 0.565score
deep-explanation
Tam Nguyen
Explain any topic from first principles as a layered, causal explanation that fully answers the question.
SkillAFreshness: AClaude Code +9 - 0.565score
orwell-writing
Tam Nguyen
Use when an agent is asked to draft, rewrite, edit, review, polish, copyedit, simplify, humanize, or create written prose, including creative writing, essays…
SkillAFreshness: AClaude Code +9 - 0.565score
document-review
Omni
Review a draft document for clarity, structure, tone, and errors, and return prioritised feedback.
SkillAFreshness: AClaude Code +9 - 0.565score
job-description
Omni
Write a job description in house format from a hiring manager's notes.
SkillAFreshness: AClaude Code +9 - 0.565score
linkedin-post-writing
Omni
Write, improve or audit a LinkedIn post, or draft hook options, with a reader-first method built to grow followers.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 19, 2026 (11 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 1.0.0
- License
- MIT
- Repository
- github.com/amahony/agent-templates
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Tam Nguyen
- 0.570score
paper-opportunity-radar
Tam Nguyen
Run a cumulative daily or one-off sweep of research papers on a chosen topic, audit their claims, methods, integrity signals and independent support, then…
SkillAFreshness: AClaude Code +9 - 0.570score
business-naming
Tam Nguyen
Generate, refine, compare and, when needed, screen distinctive names for startups, products, apps, developer tools, open-source projects, product families…
SkillAFreshness: AClaude Code +9 - 0.570score
mastery-plan
Tam Nguyen
Build an evidence-based path toward exceptional capability in a field or skill, and coach the learner along it.
SkillAFreshness: AClaude Code +9 - 0.565score
deep-explanation
Tam Nguyen
Explain any topic from first principles as a layered, causal explanation that fully answers the question.
SkillAFreshness: AClaude Code +9 - 0.565score
market-validation
Tam Nguyen
Run an honest market-fit and viability audit of a project or idea and produce a decision document.
SkillAFreshness: AClaude Code +9 - 0.565score
orwell-writing
Tam Nguyen
Use when an agent is asked to draft, rewrite, edit, review, polish, copyedit, simplify, humanize, or create written prose, including creative writing, essays…
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get blog-post-writing running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
blog-post-writing skill. Agent Mart. https://agentmart.dev/skills/amahony-blog-post-writing (last updated 2026-10-24)