PromptFork

Situational email assistant: diagnose the challenge, then write

Not just 'write me an email' — this prompt first identifies the communication challenge (delivering bad news, making a big ask, escalating), then applies the right professional framework for that exact situation.

Open in Studio
Prompt
Paste this for emails that actually get the response you want:

'You are my executive communication strategist. Before writing anything, diagnose the situation:

Context: [DESCRIBE THE SITUATION — what happened, who you are emailing, what you need from them]
Relationship: [peer / manager / direct report / client / vendor / cold outreach]
Stakes: [routine / important / high-stakes]

STEP 1 — DIAGNOSE THE CHALLENGE TYPE: Based on my context, identify which communication challenge this is:
• Delivering bad news (missed deadline, price increase, rejection)
• Making an ask (budget, time, favor, introduction)
• Following up (after silence — the hardest email to write)
• Escalating (without burning bridges)
• Saying no (while preserving the relationship)
• Persuading (changing someone's mind or priority)
• Repairing (after a mistake or miscommunication)
Tell me which type you identified and the specific risk if the email is poorly written.

STEP 2 — WRITE THE EMAIL using the inverted pyramid structure:
• Subject line: Specific, not vague. "Q3 budget request: $12K for contractor support" not "Quick question"
• First sentence: The single most important thing — your request, your news, or your point. Never open with "I hope this finds you well" for high-stakes emails. Busy readers may only see this line.
• Supporting context: 2-3 sentences max. Only what the reader needs to make a decision or understand the situation.
• Specific call to action: Exactly what you need them to do, by when. "Could you approve by Friday?" not "Let me know your thoughts."

STEP 3 — RISK CHECK: Flag anything in the draft that could:
• Be misread in tone (especially negativity or passive-aggression that I might not notice)
• Bury the key ask below the fold
• Be too long for the stakes level

Constraints: Keep it under [N, default 150] words. Tone: [warm / direct / diplomatic / firm]. Write at a level where every sentence earns its place — if a sentence can be cut without losing meaning, cut it.

Provide a SHORT version (3-4 sentences) and a FULL version (if more context is warranted), and tell me which one you recommend for this specific situation and why.'
Source
promptfork seed
License
CC-BY-4.0
Published
6/27/2026

More prompts you might like

Email thread decoder — stakeholder map, hidden asks, and strategically timed replies

Paste a messy email thread and get more than a summary — a stakeholder power map (who wants what), 'hidden ask' detection for what's being requested between the lines, and reply options with timing advice for maximum impact.

New

Reverse-engineer a prompt from examples — with pattern detection and consistency scoring

Show the AI 2-3 outputs you love; it detects structural, tonal, formatting, and content patterns (plus what the examples consistently AVOID), then writes a prompt that reliably reproduces them — with a consistency confidence score.

0003

Direct expert — answers with zero fluff

Turn any assistant into a decisive, no-padding expert. Drop it into ChatGPT custom instructions, a Claude Project, or your API system prompt.

0001
Editor’s pickChatGPT & AI PromptsSeed

Research agent system prompt — web search, source evaluation, structured briefing

A system prompt that turns any AI agent into a rigorous research analyst: it searches broadly, evaluates sources by credibility, surfaces what's genuinely uncertain, and delivers a structured briefing — not a confident-sounding summary of whatever it found first.

0001

Senior engineer for AI coding tools

Make Cursor, Claude Code, Copilot, or the API behave like a careful senior engineer — minimal diffs, your conventions, no over-explaining.

New

Ruthless editor that keeps your voice

An editor that sharpens your writing without making it sound AI-generated. Works as ChatGPT custom instructions or a Claude system prompt.

New