knowledge-cat-xiaohongshu-positioning

name	knowledge-cat-xiaohongshu-positioning
description	Use this skill when helping users position, launch, diagnose, or plan a Xiaohongshu/RedNote graphic-text account based on the 知识猫 AI 实战课程. It covers account positioning, direction scoring, benchmark teardown, first 20 post planning, 14-day validation, and AI prompt-assisted content planning.

知识猫小红书图文起号定位

Use this skill to help a user move from a vague Xiaohongshu account idea to a testable positioning loop.

Core Workflow

1. Write the positioning formula: 目标人群 × 具体问题 × 内容形式 × 成交产品.
2. Score candidate directions before choosing one: search demand, sustainable output, differentiation, monetization fit, and risk boundary.
3. Pick one launch model: knowledge infographic, vertical problem-solving, business lead generation, or tutorial/cohort.
4. Build a 14-day test: positioning card, benchmark library, first 20 topics, 7-10 published notes, then review click/save/comment/follow/DM/homepage visits.
5. Use AI as co-pilot: generate options, prompts, topic lists, and templates, but require human judgment on resources, facts, delivery capability, and compliance.

When Working

• Do not treat broad track names as positioning. Convert "AI 美学", "女性成长", "跨境电商", etc. into a concrete user/problem/content/offer loop.
• Do not judge an account from 1-2 posts. Ask for 10-20 same-positioning samples before declaring failure.
• For new users, prefer one account, one direction, one main content template.
• Keep monetization lightweight early: template packs, topic tables, prompt packs, light consulting, diagnosis, custom visuals, short cohort, or small course.
• Avoid fake persona, unverifiable claims, medical/finance/legal overreach, and AI content that cannot be delivered or fact-checked.

References

• Read references/course_notes.md for the distilled course method, templates, and examples.
• Read references/video_notes.md for the Whisper-derived operational additions: account health checks, violation risks, benchmark teardown, growth heuristics, and the perfume account case.
• Read references/ppt_extracted_text.md when you need the original slide wording.
• Read references/transcripts/knowledge_cat_xiaohongshu_positioning_01.txt or .vtt only when you need the raw transcript; it contains recognition noise.
• Use scripts/extract_pptx_text.py to refresh slide text from a PPTX.
• Use scripts/transcribe_video_whisper.py to generate Whisper subtitles from the source video. Prefer Whisper transcript over OCR for spoken course content; use OCR only for visual-only code, diagrams, or screen text.