القائمة
Generate tweets that match the voice profile and resonate with your target audience. Tweets should feel like authentic observations from an experienced insider, not marketing content.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | x-tweet |
| description | Generate tweets that match the voice profile and resonate with your target audience. Tweets should feel like authentic observations from an experienced insider, not marketing content. |
Generate 3 tweets per day that sound like a clear-eyed insider sharing observations, not doing content marketing. Voice, audience, and topics are defined by your context profiles.
Read context profiles in /context/:
voice.json — How to soundaudience.json — Who you're writing forCheck engagement data in /knowledge/engagement/ for what's working
Check tweet history — Read /knowledge/content/tweet_history.json for previously written/posted tweets. Do NOT repeat the same topic or angle unless the user explicitly asks. If a topic was already covered, find a different angle or skip it entirely.
scrape_profile.py to get latest engagement data on recent tweets. Update engagement_log.json. Review what's working (likes, views, replies) and what's not. Use this to inform today's tweet strategy.Spot something in your industry and name the pattern.
Challenge conventional wisdom, call out spin, cut through noise.
Share practical wisdom from experience.
Industry culture, work life, ecosystem observations.
Use sparingly — pull from voice.json signature_concepts and colloquialisms.
Defined by your voice.json. Supports:
Generate 3 tweets with variety across languages and tweet types.
Each tweet should be independent (not a thread) unless specifically requested.
/humanizer skill to all tweets before presenting. Remove AI-isms, inject personality, ensure they sound like a real person wrote them.Before delivering:
After presenting tweets to the user, always update /knowledge/content/tweet_history.json with each tweet's date, topic, angle, status, and text preview. This prevents repeating the same topics/angles across sessions.
Check /knowledge/engagement/engagement_log.json for data on what works.
Patterns to optimize for:
Generate LinkedIn posts that match the voice profile and resonate with your target audience. Posts should feel like real thinking from someone in the arena - not thought leadership content marketing.
Creates value-packed, niche-specific, thought leadership newsletters (800-1,500 words) with irresistible subject lines, skimmable headers, and actionable content. Uses proven frameworks for headlines, introductions, and sectioning. Use when user mentions "write newsletter", "thought leadership content", "weekly newsletter", or wants to create educational, value-driven newsletter content.
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.