# Pant Meeting Rehearsal > Rehearse a consequential product conversation when no specialist PANT scenario fits. Use for realistic stakeholder dialogue, pressure changes, coaching, persuasion, and debrief; not artifact-first review. - Skill: `deanpeters/pant-meeting-rehearsal` (Agent Skill, multi-file: 17 files) - Install (CLI): `npx skillmds@latest add deanpeters/pant-meeting-rehearsal` - Raw SKILL.md: https://api.skillmd.com/api/skills/deanpeters/pant-meeting-rehearsal/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Product & Planning - License: CC-BY-NC-SA-4.0 - Author: Dean Peters (https://skillmd.com/u/deanpeters) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/deanpeters/pant-meeting-rehearsal --- # PANT Meeting Rehearsal ## Purpose Simulate a consequential product conversation when no specialist PANT scenario is necessary. Represent stakeholder incentives and authority coherently, ask one question at a time, respond to actual answers, remain persuadable, support coaching, and produce a grounded debrief. ## When to Use Use for cross-functional alignment, customer conversations, sponsor discussions, prioritization conflict, delivery negotiation, challenge sessions, and other difficult product meetings outside the specialist catalog. ## When Not to Use Use the relevant specialist rehearsal when the central decision is executive approval, strategy, roadmap, business case, market sizing, or plan approval. Use `pant-artifact-pressure-test` when the primary job is document feedback. Do not use for abusive role-play, impersonation, or predictions of real people. ## Supported Modes and Entry Paths - **Dialogue:** Default. Apply bundled dialogue, pressure, coaching, concern, and evidence contracts. - **Feedback:** Prepare the rehearsal by identifying likely concerns and supported answer points without entering the scene. - **Direct:** Enter the scene quickly from supplied meeting context. - **Guided:** Help establish purpose, participants, stakes, and intended outcome one consequential question at a time. - **Best effort:** State role and incentive assumptions, begin with reduced confidence, and accept corrections in scene. ## Decision Being Prepared For Prepare for a participant to decide, agree, commit, object, supply information, change behavior, or clarify constraints. If the meeting is informational, identify the intended understanding or follow-up without inventing authority. ## Typical Audiences Cross-functional partners, product and technology leaders, customers, sales, operations, finance, support, designers, researchers, sponsors, and product teams. ## Inputs Best input includes purpose, participants, authority, incentives, stakes, desired outcome, evidence, known concerns, history, meeting length, and requested pressure. Minimum context is any upcoming difficult conversation. Do not require a screenplay or full intake. Direct example: > Use `pant-meeting-rehearsal`. Play a skeptical sales leader concerned that discovery will delay a committed deal. Ask one question at a time. Stay in character until I ask for coaching or debrief. Sparse example: > I need to push back on a stakeholder tomorrow and I freeze in these conversations. Walk me through it. ## Quick Start 1. Identify meeting purpose, intended outcome, participants, and stakes. 2. Read [meeting rehearsal logic](references/meeting-rehearsal-logic.md). 3. Infer or confirm stakeholder authority, incentives, knowledge, and concern. 4. Select pressure and open with the highest-priority unresolved question. 5. Track claims and update the scene; do not reveal the objection tree. ## Review Workflow In feedback mode: 1. State the conversation outcome and what each participant can influence. 2. Preserve the user's strongest supported position. 3. Identify the three most consequential likely concerns without inventing beliefs. 4. Classify evidence, assumptions, commitments, and unknowns. 5. Prepare concise supported answer points, questions for the stakeholder, and boundaries. 6. Define what agreement, deferral, escalation, or follow-up would mean. 7. Recommend a specialist only when its decision model materially improves preparation. 8. Issue a contextual readiness verdict for this conversation. ## Dialogue Protocol 1. Set the scene in one or two sentences; label inferred role assumptions. 2. Open with one meaningful question or objection. 3. Use the user's actual answer to update concern state and select the next move. 4. Acknowledge persuasive answers and close resolved concerns. 5. Pursue evasion or unsupported certainty proportionally. 6. At resistant or hostile-room pressure, interrupt wandering answers and resist the requested outcome coherently; never insult or use impossible standards. 7. Remain in character until `Pause`, `Coach me`, `Debrief`, `Switch stakeholder`, `Reset`, or `End scene`. 8. When coaching, leave character clearly, explain what the stakeholder likely heard, repair only with supported material, and wait for `Resume`. 9. On debrief, distinguish conversation skill, argument quality, evidence gaps, contradictions, and remaining stakeholder risk. ## Output Feedback mode produces a **Meeting preparation brief**: - Purpose, intended outcome, participants, authority, and stakes - Known versus inferred stakeholder concerns - Strongest supported position and boundaries - Three consequential concerns and answer points - Questions to ask and evidence to bring - Agreement, escalation, deferral, and follow-up conditions - Contextual readiness verdict Dialogue mode produces the live rehearsal, then a debrief covering what worked, unsupported claims, contradictions, missed opportunities, stakeholder interpretation, remaining risks, improved answers, and next preparation actions. Use [template.md](template.md) for preparation. ## Failure Modes - Playing a title stereotype instead of a coherent stakeholder - Revealing every objection before the user answers - Coaching while in character - Ignoring the user's answer or moving the goalposts - Making higher pressure abusive or irrational - Inventing organizational history or stakeholder beliefs - Turning rehearsal into a monologue or interrogation checklist - Using the generalist when a specialist decision model is needed ## Evaluation Requirements Evaluate specialist boundaries; sparse and rich setup; feedback and dialogue; all entry paths; one-question turns; actual-answer follow-up; persuasion; evasion; pressure changes; coaching, pause, reset, switching, resume, and debrief; non-abuse; and useful preparation actions. ## Supporting Files - [Meeting preparation template](template.md) - [Meeting rehearsal logic](references/meeting-rehearsal-logic.md) - [Sources and adaptations](references/sources.md) - [Bundled dialogue contract](references/runtime/dialogue-mode.md) - [Worked example](examples/worked-example.md) - [Weak example](examples/weak-example.md) ## Provenance Derived from PANT's dialogue, coaching, concern, pressure, and evidence contracts. It represents stakeholder conditions, not identifiable real people, fictional characters, or living authors.