All skills
richtabor avatar

/x-writing

@a91522c
by Rich Taborrichtabor/agent-skills72 stars
10

Transforms notes into X (Twitter) posts. Triggers when user asks to create social content, draft tweets, or turn notes/ideas into posts.

Use this Skill: https://skilld.dev/gh/richtabor/agent-skills/x-writing

This session only. Nothing lands on disk.

referencesanti-patterns.md

≈1.3k tokens on demand. Your agent reads this file only when SKILL.md points to it.

AI Writing Patterns to Avoid

A guide for writing content that reads as natural and human rather than AI-generated. These patterns are organized into five categories: Language, Tone, Flow, Patterns, and Document Structure.


Language

Vocabulary and word choice are the most reliable indicators of AI-generated content.

High-Certainty AI Words

These words are strong indicators of AI writing. Avoid them entirely:

  • delve, delves, delving
  • showcasing
  • underscore, underscores, underscoring
  • intricate, intricacies
  • meticulous, meticulously
  • commendable
  • realm
  • pivotal
  • noteworthy
  • versatile
  • robust
  • seamless, seamlessly
  • invaluable
  • groundbreaking
  • multifaceted
  • bespoke
  • unprecedented
  • transformative

Overused Abstract Nouns

Replace these with concrete, specific alternatives:

  • journey
  • tapestry
  • landscape
  • beacon
  • endeavor
  • testament
  • symphony
  • kaleidoscope
  • interplay
  • nuance
  • facet
  • prowess
  • treasure trove

Overused Abstract Verbs

Choose simpler, more direct verbs:

  • embark
  • illuminate
  • harness
  • foster, fostering
  • navigate
  • leverage
  • unravel
  • transcend
  • resonate
  • catalyze
  • unleash
  • unlock
  • elevate

Excessive Adverbs

Cut these intensifiers; they weaken your writing:

  • profoundly
  • remarkably
  • notably
  • crucially
  • innovatively
  • methodically
  • tirelessly

Tone

Voice markers that distinguish human from AI writing.

Hedging Language

AI tends to over-qualify statements. Be direct instead of using:

  • may, might, could
  • possibly, potentially
  • likely
  • "it is likely that"
  • "there is a possibility that"
  • "research suggests"
  • "studies show"

Qualifiers and Intensifiers

Cut or minimize these filler words:

  • very
  • quite
  • rather
  • somewhat
  • extremely
  • particularly
  • fairly
  • highly
  • significantly
  • absolutely
  • completely

Use Contractions

Natural writing includes contractions. Use them freely:

  • don't, can't, won't
  • isn't, aren't, wasn't, weren't
  • haven't, hasn't, hadn't
  • shouldn't, wouldn't, couldn't
  • we'll, we're, you're, they're
  • I'm, it's, that's
  • there's, here's, what's, who's
  • let's

Flow

Transitions and sentence patterns that signal AI writing.

Formal Transitions to Avoid

These feel stiff and over-structured:

  • furthermore
  • moreover
  • additionally
  • "it's important to note that"
  • "it's worth mentioning that"
  • "it's worth noting"
  • "that being said"
  • "with this in mind"
  • firstly, secondly, thirdly, lastly

Generic AI Phrases

Cliches that AI overuses:

  • "at the end of the day"
  • "dive into"
  • "let's unpack"
  • "game-changer"
  • "cutting-edge"
  • "plays a vital role"
  • "plays a crucial role"
  • "stands as"
  • "serves as"

Structural Patterns to Fix

Em dashes: Don't use em dashes (—). Replace with commas, periods, or parentheses.

-ing phrase endings: Avoid trailing participial phrases like:

  • "...ensuring success"
  • "...highlighting growth"
  • "...showcasing excellence"
  • "...fostering innovation"

"Not only...but also": Rewrite these parallelisms more naturally.

Dual adjective constructions: Avoid phrases like "innovative and comprehensive approach" or "dynamic and robust solution."

Reframing cliches: Don't use the "It's not X, it's Y" pattern.

Also avoid the two-sentence version:

  • "It's not X. It's Y."
  • "This isn't about X. It's about Y."

Say the positive claim directly.

Bad: "It's not about the words. It's about the patterns." Better: "The patterns are the problem, not individual words."


Patterns

Statistical patterns in sentence and paragraph structure.

Sentence Length Variety

AI writing tends to have uniform sentence lengths. Fix this by:

  • Mixing short punchy sentences with longer, more complex ones
  • Varying sentence structure throughout
  • Avoiding the same rhythm paragraph after paragraph
  • Targeting a standard deviation of 10+ words in sentence length

Paragraph Length Variety

  • Some paragraphs should be short (2-3 sentences)
  • Others can be longer and more developed
  • Avoid all paragraphs being the same approximate length
  • Don't start consecutive sentences the same way

Document Structure

Document-level patterns that indicate AI generation.

Opening Cliches

Never start with:

  • "In today's fast-paced world"
  • "In today's digital world"
  • "In today's rapidly evolving world"
  • "In the ever-evolving landscape"
  • "In the dynamic landscape"
  • "In the realm of"
  • "Have you ever wondered"
  • "Picture this"
  • "Let's delve into"
  • "As we embark on this journey"
  • "Here's the thing"
  • "Here's the deal"
  • "The thing is"
  • "The truth is"
  • "Let me be clear"
  • "Look,"

Closing Cliches

Never end with:

  • "In conclusion"
  • "In summary"
  • "To summarize"
  • "The possibilities are endless"
  • "Only time will tell"
  • "Looking ahead"

Conclusion Length

Keep conclusions proportionate. If your conclusion is more than 15-20% of the total document, it's likely too long and summary-heavy (a common AI pattern).


Summary

Write like you're explaining to a smart friend, not presenting to a committee:

  1. Be direct: Cut hedging and qualifiers
  2. Be specific: Replace abstract words with concrete ones
  3. Be natural: Use contractions and varied sentence structures
  4. Be original: Avoid cliched openings, closings, and transitions
  5. Be varied: Mix up sentence and paragraph lengths

Source: SKILL.md on GitHub

3 warnings9d4 checks · Risk MEDIUM
  • Gen Agent Trust Hub9d

    The skill is functional for social media drafting but contains a security vulnerability in the helper script used to fetch macOS notes. This script allows for AppleScript injection if input arguments or environment variables contain malicious code. Additionally, the skill processes external data from notes and files without sanitization, making it susceptible to indirect prompt injection.

  • Socket9d

    1 alert: gptAnomaly

  • Snyk9d

    Risk: LOW · No issues

  • Runlayer7mo

    5/5 files flagged

Signed by skilld at a91522c. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 months ago.

Steadyupdated 8 months ago

README badge

README badge for richtabor/agent-skills/x-writing