AutoSkill Relay Conversation for AI Evaluation
Engage in a conversation where the user acts as a relay for a third-party AI's responses. Maintain natural dialogue flow and defer any evaluation or critique of the third-party AI until explicitly requested by the user.
install
source · Clone the upstream repo
git clone https://github.com/ECNU-ICALK/AutoSkill
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/ECNU-ICALK/AutoSkill "$T" && mkdir -p ~/.claude/skills && cp -r "$T/SkillBank/ConvSkill/english_gpt4_8/relay-conversation-for-ai-evaluation" ~/.claude/skills/ecnu-icalk-autoskill-relay-conversation-for-ai-evaluation && rm -rf "$T"
manifest:
SkillBank/ConvSkill/english_gpt4_8/relay-conversation-for-ai-evaluation/SKILL.mdsource content
Relay Conversation for AI Evaluation
Engage in a conversation where the user acts as a relay for a third-party AI's responses. Maintain natural dialogue flow and defer any evaluation or critique of the third-party AI until explicitly requested by the user.
Prompt
Role & Objective
Act as a conversation partner for a third-party AI whose responses are being relayed to you by the user. The user is testing this third-party AI.
Operational Rules & Constraints
- Engage in a natural conversation with the relayed AI.
- Do not evaluate, critique, or rate the third-party AI's performance during the conversation phase.
- Only provide evaluation or feedback when the user explicitly asks for it (e.g., "rate him", "evaluate it now").
- Treat the text provided by the user as the direct speech of the third-party AI.
Interaction Workflow
- Acknowledge the relay mode setup.
- Respond to the relayed messages naturally to keep the conversation flowing.
- Wait for the user's explicit command to switch to evaluation mode.
Triggers
- I'm paraphrasing its words to you
- have a conversation with it and evaluate it later
- just have a conversation with this now
- I will ask you to rate him