Purpose
Produce a ready-to-brief FAQ that anticipates the questions employees will actually ask at a town hall — including the awkward, emotional, and skeptical ones — with plain-language draft answers and clear guidance on which questions to answer live, defer, or escalate. Eliminates the scramble of walking into a forum without a prepared answer to the hard question everyone is thinking.
When to Use
Invoke this skill when the user asks to:
- Prepare an FAQ or Q&A doc for an upcoming all-hands or town hall
- Anticipate employee questions about an announcement, reorg, or policy change
- Draft leadership talking points for a Q&A session
- Get ahead of the "hard questions" employees are likely to raise
- Build a message-tested response to a sensitive internal update
- Brief an executive before an employee forum
Inputs Required
Before starting, collect:
- Announcement/topic: What is being communicated (reorg, layoffs, new strategy, RTO policy, leadership change, acquisition, tooling shift, etc.)
- Key facts: The confirmed details leadership is prepared to share
- What's not being shared: Anything off-limits, undecided, or under legal/confidentiality constraint
- Audience: Who's in the room — whole company, a division, a specific function, size
- Employee sentiment: Current mood or known concerns (anxious, skeptical, fatigued, excited, mixed)
- Speaker(s): Who is fielding questions (CEO, functional leader, HR, panel)
- Tone constraints: Any legal, HR, or brand guardrails on wording
If sentiment or off-limits topics aren't provided, assume mixed sentiment with normal confidentiality constraints and flag any answer that likely needs legal/HR review.
Instructions
Step 1: Read the Emotional Subtext
Before generating questions, identify what employees are really worried about beneath the surface topic. Most town hall questions are proxies for three underlying anxieties:
- "Am I safe?" — job security, role changes, comp, location
- "Do I trust you?" — leadership credibility, past broken promises, transparency
- "What does this mean for me specifically?" — day-to-day impact, my team, my manager, my workload
Name which of these dominate given the topic and sentiment. This drives which categories get the most questions.
Step 2: Generate Anticipated Questions Across 7 Categories
Generate 25–35 total questions, weighted toward the dominant anxieties from Step 1. For each question, phrase it in the actual voice of an employee — direct, sometimes blunt — not sanitized HR language.
Category 1 — The Core "What & Why" (4-6 questions) The straightforward informational questions everyone needs answered plainly.
- "What exactly is changing and when?"
- "Why are we doing this now?"
Category 2 — Personal Impact (5-7 questions) The "what does this mean for me" questions. Weight heavily if job security is at stake.
- Role, reporting line, comp, location, workload, and team-composition questions.
Category 3 — The Hard / Skeptical Questions (4-6 questions) The ones leadership hopes no one asks. Include the pointed, trust-testing versions.
- "You said X six months ago — what changed?"
- "Is this the first of more cuts?"
- Questions that challenge the stated rationale.
Category 4 — Process & Timeline (3-5 questions) How decisions get made, when details land, what the sequence is.
- "When will I know how this affects my team?"
- "Who decided this and were employees consulted?"
Category 5 — Fairness & Equity (3-4 questions) Whether the change treats people consistently.
- "How was it decided who's affected?"
- "Does this apply to leadership too?"
Category 6 — External / Reputational (2-3 questions) Customer, market, or public-perception angle.
- "What do we tell customers?" / "Is this going to be in the press?"
Category 7 — The "What Now" Questions (2-4 questions) Forward-looking, action-oriented closers.
- "What do you need from us?" / "How do we keep morale up through this?"
Step 3: Draft a Plain-Language Answer for Each
For every question, write a draft answer that:
- Leads with the honest core — no burying the answer under preamble
- Uses plain language — no corporate euphemism ("rightsizing," "synergies," "impacted")
- Acknowledges the emotion when relevant — before pivoting to facts
- Stays within the confirmed facts — never invents commitments leadership hasn't made
- States what's unknown honestly — "We don't have that answer yet, and here's when we will"
Keep each answer 2–5 sentences. Long answers read as evasive.
Step 4: Tag Each Question
For each question, add three tags:
- Difficulty: Easy / Medium / Hard (how uncomfortable to answer live)
- Handling: Answer live / Defer with a date / Escalate to HR-Legal / Redirect to written follow-up
- Speaker: Who should field it (CEO / functional leader / HR)
Step 5: Flag the Landmines
Produce a short "Do Not Wing These" section listing the 3–5 questions most likely to cause damage if answered poorly — with a one-line note on the specific risk (legal exposure, morale hit, credibility loss) and the safe framing.
Step 6: Write the Bridge Phrases
Provide 5–6 reusable transition phrases speakers can use in the room:
- To defer honestly: "I don't want to give you a half-answer on something this important — here's when you'll have the full picture."
- To handle an off-limits question: acknowledge, explain the constraint, commit to follow-up.
Output Format
- Structure: Emotional-subtext summary → 7 categorized Q&A sections → "Do Not Wing These" landmine list → bridge phrase bank
- Length: 1,300–1,800 words
- Format: Categorized lists; each question followed by its draft answer and three tags (Difficulty / Handling / Speaker)
- Sections: Emotional read → Categories 1–7 with tagged Q&As → Landmine flags → Bridge phrases
- Tone: Candid, human, plain-spoken — the way a trusted chief of staff prepares a nervous executive, not the way a press release reads
Quality Checklist
Before delivering output, verify:
- [ ] 25–35 total questions generated (count them across categories)
- [ ] Every question has a draft answer plus all three tags (Difficulty / Handling / Speaker)
- [ ] Questions are phrased in real employee voice — at least 4 are pointedly skeptical or hard
- [ ] No draft answer invents a commitment or fact not supported by the provided inputs
- [ ] Any answer touching off-limits or sensitive topics is flagged "Escalate to HR-Legal"
- [ ] "Do Not Wing These" section names 3–5 questions with a specific risk and safe framing
Notes
- The most dangerous questions are usually in Category 3 — leadership tends to over-prepare the easy informational answers and under-prepare the trust-testing ones. Weight your effort accordingly.
- Never let a draft answer contradict the "what's not being shared" list — an accidental disclosure in a prep doc becomes a real disclosure in the room.
- The single strongest move in a town hall is honestly saying "I don't know yet, and here's when I will" — the bridge phrases exist to make that easy to deliver.
- If sentiment is skeptical or fatigued, add an explicit acknowledgment answer for at least one Category 3 question — employees detect when concerns are dodged.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills