hexiecshexiecsCommunity·1 Agent Skills Included

talk-normal

Make any AI answer directly without filler or corporate fluff

Installs a curated system prompt that makes any LLM answer directly, cutting verbose output by 56-73% while keeping all useful information. Eliminates filler phrases, throat-clearing openings, summary-stamp closings, and padded corporate tone in both English and Chinese. Injects the rules into your agent's config file once, so every future reply stays concise with zero per-message effort.
npx skills add hexiecs/talk-normal --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install talk-normal?

Run `npx skills add hexiecs/talk-normal --all -g -y` in your terminal to install the skill globally, then run the included install script to inject the rules into your agent config.

How to stop AI from writing verbose answers?

talk-normal adds a curated system prompt to your agent that bans filler phrases, restating the question, and summary closings, cutting response length by up to 73%.

Does talk-normal work with ChatGPT, Claude, and Gemini?

Yes. It works with any model that honors system prompts, including GPT, Claude, Gemini, Grok, Qwen, and DeepSeek, in both English and Chinese.

Will shorter AI answers lose important information?

No. Benchmarks across 10 test prompts show the rules preserve all useful information while removing only padding and corporate tone.

Can I remove talk-normal after installing?

Yes. Run the included install script with the uninstall flag and it cleanly removes the rule block from your config file without touching anything else.

Related Repositories in Content & Communication

View All in Content & Communication