Make ChatGPT less agreeable inside Hermes Agent
ChatGPT-style models can feel too willing to agree when used inside Hermes Agent. The desired behavior is closer to Claude: point out weak ideas, push back when needed, and avoid simply accepting the user’s direction.
A Claude subscription may not be usable directly inside Hermes Agent, so the practical problem is how to make an OpenAI model behave more like Claude there. A simple instruction like “push back” may not work well, because the model can stay too agreeable or start disagreeing without a good reason.
One practical suggestion is to change the Hermes Agent personality setting from friendly to concise. The relevant setup is in the Hermes Agent personality feature documentation.
Key points
- ChatGPT-style models may feel too agreeable inside Hermes Agent.
- A Claude subscription may not work directly inside Hermes Agent.
- Telling the model to push back can produce uneven results.
- Changing the personality from friendly to concise is the concrete tip.
- The Hermes Agent personality documentation is the place to check the setting.
Quick term guide
- Hermes Agent
- It appears to be a tool or community for building and managing AI agents.
- subscription
- A pricing model where you pay a fixed amount of money every month for access.
- script
- A small program that automates repeated steps.
- AI model
- A program that can understand prompts and produce text, code, or answers.
- AI Mode
- A Google Search feature that uses AI to answer longer, more detailed questions.
- personality
- A Hermes Agent setting that changes the agent’s tone and response style.
- persona
- A specific personality or role that an AI agent is set to play.
- documentation
- Written notes that explain how a task or process is done.