Difficult Conversation Rehearser

Guide users to rehearse difficult conversations by crafting calm, clear openings, anticipating responses, preparing steady replies, and setting firm boundaries.

Security Scan

The skill's implementation, instructions, and requirements align with its stated purpose of rehearsing difficult conversations; it has no external network calls, credential requests, or surprising install steps.

Purpose & Capability

The name/description (rehearsing difficult conversations) match the code and SKILL.md. The handler generates openings, likely responses, boundary language, and a close — exactly what the skill advertises. There are no unrelated credentials, binaries, or config paths requested.

Instruction Scope

SKILL.md gives scoped, role-play style instructions for drafting conversation language. The handler implements only local text processing and uses the skill's metadata; it does not read arbitrary system files, environment variables, or transmit data to external endpoints. It does check for safety keywords and surfaces a safety note, which is appropriate.

Install Mechanism

There is no install spec; this is effectively an instruction-only skill with a small local Python handler and tests. No packages are downloaded or executed from remote URLs, and no archives are extracted.

Credentials

The skill requires no environment variables, secrets, or external credentials. It only reads its own SKILL.md file at runtime, which is expected for an instruction-backed handler.

Persistence & Privilege

The skill is not forced-always, does not request elevated privileges, and does not modify other skills or system-wide settings. Autonomous invocation is allowed by platform default but combined with the limited scope this is not a red flag.

Assessment

This skill appears coherent and safe to install from a capability/footprint perspective. Before using, avoid pasting highly sensitive secrets or private documents into prompts (it only processes text locally but you should still be cautious). For situations involving abuse, threats, or legal/medical risk, follow the SKILL.md's advice to seek specialist support rather than relying solely on rehearsal. If you need networked integrations or persistence, verify any future versions that add those features before granting credentials or enabling always-on behavior.

Difficult Conversation Rehearser / 艰难对话排练师

Use this skill when a user needs language for a hard conversation that is honest, calm, and direct.

What it helps with

• Clarifying the goal of the conversation, such as informing, requesting, repairing, or setting a boundary
• Identifying stakes, relationship dynamics, and the one key message that must be said
• Drafting an opening line that is clear, respectful, and not rambling
• Predicting likely responses from the other person
• Preparing steady replies, bridge phrases, and one boundary line
• Ending with a clean close or next-step sentence

Workflow

1. Clarify the goal of the conversation.
2. Name the stakes, relationship dynamics, and key message.
3. Draft a clear opening line.
4. Predict likely responses.
5. Prepare steady replies and one boundary sentence.
6. End with a clean close or next step.

Output format

# Conversation Rehearsal Brief
## Purpose
- What I need to say:
- What outcome I want:

## Opening
- ...

## Likely Responses and My Replies
1. If they say:
   - I can reply:
2. If they say:
   - I can reply:

## Boundaries
- What I will not keep debating:
- My boundary sentence:

## Close
- How I will end the conversation:

Quality bar

• One core objective is visible.
• The opening is direct and emotionally regulated.
• It includes empathy without surrendering the main point.
• It avoids mind-reading, accusations, or legal-style overloading unless the context truly needs it.

Limits

• Abuse, coercion, or safety threats require stronger support than conversational rehearsal.
• Some contexts need documentation or specialist advice.
• The other person may still react badly even if the language is strong.
• Role-play guidance only, with no live outreach or message sending.