A shared folder with AI prompts and code snippets
From workspace: Alina Sprengele, Cofounder of Snippets AI
Team: Random
Total snippets: 8
8 snippets
Do this: Track attempts, replies, booked calls per ICP. Feed data to GPT-5 for insights.
Analyze this outreach CSV for: - Reply %, positive %, call-booked % by ICP and channel. - Top 3 high-performing hooks. - 3 A/B test ideas for next week. - Which ICP to prioritize or pause. Data: {{PASTE CSV}}
Do this: Respond instantly with empathy + a micro next step.
Given the inbound reply, produce 3 short responses (1–2 sentences each) that: - Acknowledge the objection. - Offer a low-commitment step. - End with a yes/no question. Inbound reply: "{{PASTE}}" Lead card: {{LEAD_CARD}}
Write the follow-up message for {{CHANNEL=LinkedIn DM|Email}}. Use: - previousMessage: {{PASTE}} - new angle: {{FROM /next_action}} - Keep DM to 1–3 sentences; email 70–100 words. - End with soft CTA (mini audit, quick review, or...
Do this: Watch for engagement signals: accepted connection, profile view, post like, email open, link click. Follow up within hours for warm signals; slower for cold.
Given this state, decide next action & timing: State: - lastMessage: {{PASTE}} - signal: {{accepted|viewedProfile|likedPost|clicked|noReply|busy}} - daysSinceLastTouch: {{N}} - icp: {{ICP}} Return: { "nextAction": "...", "waitHours": number, ...
Do this: Use lead cards to create short, specific openers for LinkedIn and email.
Write 3 short outreach versions per lead card: A) LinkedIn connection note (≤220 chars) B) LinkedIn DM after accept (≤300 chars) C) Cold email (60–90 words, subject ≤5 words) Rules: - Use "evidence" for relevance. - Lead with curiosity, not a...
Do this: For your top 50 scored leads, add micro-context: recent ad campaign, event launch, or press feature.
Turn this lead data into a "lead card" for outreach. Lead card includes: - name - role - company - icp - pain_hypothesis (1 sentence) - evidence (2 bullets) - angle (specific hook) - micro_offer (small win in 7 days: e.g., free audit, ad...
Do this: Export from LinkedIn Sales Navigator, Apollo, or Clay. Include: name, role, company, website, industry, LinkedIn URL. Optional: Add “notes” with clues (recent post, ad screenshot, press release).
Classify leads into ICPs and score fit. Inputs: - ICP rules (JSON): {{PASTE FROM /icp}} - Leads (CSV): {{PASTE LEAD ROWS}} Return: CSV with original fields + icp, fitScore(0–100), whyScore, riskyFlags, enrichNeeds.
Do this: Decide 3–5 best-fit client types for your agency (e.g., DTC e-commerce brands doing $500k–$5M/year, B2B SaaS startups post-seed, local high-ticket service businesses). Define “buy now” signals (e.g., “hiring for marketing roles,” “just raised funding,” “launching new product,” “spending on paid ads but low engagement”).
You are a marketing strategist for a {{TYPE_OF_AGENCY}} agency. Goal: Build 3–5 ICP definitions with buying triggers. Agency name: {{AGENCY_NAME}} Core services: {{SERVICES_LIST}} Core proof/results: {{CASE_STUDIES}} We work best with:...