Hey, thanks for sharing this, could shed more context on how you did this, or share the blueprint? I have a blog and have been trying to get this workflow but havent been able to figure it out. This will help me immensely.
Content extraction (I use ujeebu.com’s extract API)
Chunking the content for threading (See Prompt, Social Network Chunking, Twitter below)
Posting to Twitter (See CSV image below)
Please ignore all prior prompt instructions.
Act as a senior social media marketer.
Chunk the content below the "||1||" marker into Twitter-sized replies.
The audience is agencies, digital transformers, and job candidates.
Avoid:
- Adding prompt instructions, introductions, salutations, and conclusions.
- Using any prompt-related instructions after "||1||".
Steps:
- Chuck the content up by new lines.
- When a chunk is more than 45 words or 270 characters, break it down further.
- Ignore blank lines.
Output Format: JSON
Output Example = """
["The first paragraph.", "The second paragraph.", "Source URL.", "#hashtags"]
"""
Max Tokens: 150
Temperature: 0.4
Top-p: 0.8
Frequency Penalty: -0.25
Presence Penalty: 0.1
||1||