Paperclip Review Loop

Paperclip should review each live post in three passes: immediate signal check, same-day sentiment read, and next-day amplification decision. That keeps the content loop human-led while still using agents for discipline and pattern recognition.

Review Windows

• T+30 minutes: Check early replies, target-account engagement, and whether the hook landed with the intended audience.
• T+6 hours: Review sentiment clusters, repost quality, and whether a follow-up reply or quote post is warranted.
• T+24 hours: Log which signals mattered and whether the post should become a thread, memo, or outreach touchpoint.

Signals To Track

• Target-account repost: A founder, operator, or respected builder reposts or quotes the post.
• Save-worthy comment behavior: Comments ask for process detail or cite the post in an operator context.
• Qualified reply chain: A conversation forms around the core idea instead of drifting into generic reaction.

Escalation Thresholds

• Three or more target-account interactions: Draft a same-day follow-up post or thread while the topic remains active.
• Repeated proof requests: Promote a case-study fragment or systems breakdown within 24 hours.

Agent Roles

• Sentiment reviewer: Reads replies and quote posts for tone, misunderstanding risk, and unexpectedly strong resonance.
• Signal mapper: Flags target accounts, repost clusters, and the posts most likely to unlock introductions.
• Follow-up planner: Suggests whether the next move should be a reply, quote post, memo, or direct outreach.

Escalation Rules

• High-intent lead signal: If a founder, product lead, or operator asks for process detail, move the reply toward a direct systems review conversation.
• Misread or skepticism: Answer publicly once with concrete language, then stop if the thread turns performative.
• Unexpected traction: If a post attracts multiple target-account reposts or comments, create a same-day follow-up while the topic is still hot.