Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Blog Post Skill — measured AI agent skill | DecimalAI

nexu-io/blog-post

NewSkillScoreMaintained

A long-form article / blog post — masthead, hero image placeholder, article body with figures and pull quotes, author byline, related posts. Use when the brief asks for "blog", "article", "post", "essay", or "case study".

by nexu-io

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Frontend / UI pack

decimalai skills pull nexu-io-blog-post

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/nexu-io-blog-post/SKILL.md

Raw file: /s/nexu-io-blog-post/SKILL.md ↗ · machine-readable summary at /s/nexu-io-blog-post.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 7d ago

No — -50% lift+20% tokens0% turns50% of whole cases pass with skill (11/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	10%	0%
case-02	✗→✓	▲ Improved	10%	0%
case-08	✗→✓	▲ Improved	10%	0%
case-16	✓→✗	▼ Worse	10%	0%
case-05	✓→✗	▼ Worse	10%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

nexu-io / open-design

design-templates/blog-post/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/8/2026Version v1

Skill contents

SKILL.md

Blog Post Skill

Produce a single long-form article page — editorial layout, no chrome.

Workflow

Read the active DESIGN.md (injected above). Lean into the typography

tokens — long-form is 70% type, 20% image, 10% chrome.

Pick the topic from the brief and write a real article — at least 600

words across 4–6 H2 sections. No lorem ipsum.

Sections, in order:

Masthead — small wordmark + 4–6 nav links, plain.
Article header — category eyebrow, headline (display token, large),

deck (1–2 sentence subhead), author name + role + date.

Hero image — a 16:9 placeholder block using a DS-tinted gradient or

solid fill (no external images). Add a 1-line caption underneath.

Body — alternating prose paragraphs with at least:
1 pull quote (large display type, accent rule on the inline-start edge so the layout flips correctly under dir="rtl").
1 figure (image placeholder + caption).
1 list (numbered or bulleted).
1 inline blockquote.
Author footer — author avatar (initials in a circle), bio paragraph.
Related — 3 cards linking to other posts. Each card: tiny image

block, title, 1-line excerpt, date.

Write a single HTML document:

<!doctype html> through </html>, CSS inline.
Article body uses the DS body font, centered, max-width per DS layout

rule (typically 680–720px).

Drop caps (first-letter) only if the DS mood is editorial / serif —

skip on tech-y DSes.

data-od-id on the headline, hero, body, pull quote, related grid.

Self-check:

Type hierarchy is unambiguous — H1 is clearly the headline; H2s are

section dividers; pull quotes do not compete with H1.

Line length 60–75 chars for body prose.
Accent appears at most twice (eyebrow + pull-quote rule, or one link).
The page reads like a magazine, not a marketing landing.

Output contract

Emit between <artifact> tags:

<artifact identifier="post-slug" type="text/html" title="Article Title">
<!doctype html>
<html>...</html>
</artifact>

One sentence before the artifact, nothing after.

Want to see if nexu-io/blog-post would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.