0
votes
0
resp
4
views
Generates the customer facing escalation message and the internal note that briefs the manager at the same time, so the customer never has to repeat their issue. Source: https://su
9 shown
0
votes
0
resp
3
views
Acts as an instructional coach that asks about your topic, grade level, prior knowledge, and learning goals, then drafts a multi modality lesson plan and refines it with you around
0
votes
0
resp
2
views
Turns confusing lab report numbers into a plain language explanation, flags patterns and possible medication interactions worth asking about, and generates questions for your next
0
votes
0
resp
4
views
A structured response template that enforces acknowledgment first replies, bans hollow phrases like 'I understand your frustration,' and caps responses at 100 words. Use it to gene
0
votes
0
resp
6
views
Turns any AI into a patient Socratic tutor that probes the student's level, asks leading questions instead of stating answers, and confirms genuine understanding before moving on.
0
votes
0
resp
4
views
A fill in the blank system prompt that gives any customer support bot its role, tone, and safety constraints in one drop in block. Replace three placeholders and your bot is ready
0
votes
0
resp
18
views
From Chapter 9 of Anthropic's interactive prompt engineering tutorial. A canonical complex prompt skeleton that combines task context, tone, rules, examples, conversational history
chatbottemplatesystem-promptxml-structuredanthropicClaude Opus 4.7Claude Sonnet 4.6Claude Haiku 4.5Claude 3.5 Sonnet
0
votes
0
resp
17
views
Opening sections of Anthropic's officially published system prompt for Claude Sonnet 4.5 release notes, Sept 2025 . Notable for prose first formatting guidance, sparse bullet usage
0
votes
0
resp
32
views
The most pulled RAG prompt on LangChain Hub. Three sentence answer cap plus an explicit "say you don't know" clause makes it a safe default for production RAG. Source: https://smit