Difficult Conversation Prep
SkillDev toolsPrepare for challenging conversations using Prepare-Listen-Empathize-Clarify-Solve framework.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Difficult Conversation Prep skill
What this skill tells your AI
The instructions your AI receives, as published by itseffi/agentic-os in .agents/skills/difficult-conversations/SKILL.md and read by ahel’s review.
Prepare for challenging work conversations using a structured framework.
When to Use
Before any conversation you're dreading or that has conflict potential.
The Framework: Prepare-Listen-Empathize-Clarify-Solve
1. Preparation Notes
Document:
- Your intent (what you want to achieve)
- Non-negotiables (what you won't compromise on)
- Specific facts/examples (concrete, not accusations)
- Language traps to avoid
- Desired outcome
- Success criteria
- Medium choice (in-person/video/async)
2. Opening Lines (3-5 options)
Create openers that are:
- Neutral, non-accusatory
- Reference concrete impact
- Invite their perspective
Examples:
- "I wanted to talk about [situation] because I've noticed [impact]..."
- "I'd like to understand your perspective on [issue]..."
- "Can we discuss [topic]? I want to make sure we're aligned..."
3. Questions to Ask (5-7)
Use open questions (what/how/when):
- Avoid "why" framing (feels accusatory)
- Include one perspective-taking question
- Include one constraints question
4. Anticipate Reactions
Prepare responses for:
- Denial: [De-escalating response]
- Deflection: [Redirect response]
- Emotional reaction: [Calming response]
- Counter-accusation: [Non-defensive response]
5. Clarify and Solve
- State shared goals
- Present options with trade-offs
- Agree on commitments
- Assign owner and timeline
- Schedule follow-up
Self-Management Checklist
- Deep breaths before starting
- Slow speaking pace
- Use pauses deliberately
- Notice body tension
- Stay curious, not defensive
Output
- 1-page prep doc
- Opening lines (3-5)
- Questions list
- Reactions/responses table
- Close and next steps
- Escalation flags (if HR/compliance needed)
When Not to Use
Do not use this skill when the request is unrelated, low-stakes, or better handled by a simpler direct response.
Signals
- GitHub stars
- 112
- Forks
- 21
- Last commit
- Aug 2026
Advanced
- Catalog kind
- skill
- Gateway key
difficult-conversations- Source
- github.com/itseffi/agentic-os