anti-rlhf

Here are 2 public repositories matching this topic...

frangelbarrera / code-review-agent-mcp

AI code review agent MCP server. Reviews code like a kernel maintainer: blunt, technical, no sugarcoating. Detects bugs + OWASP Top 10 security vulnerabilities. 4 harshness levels. Anti-RLHF.

Updated Sep 24, 2026
Python

5ynthaire / 5YN-ScareQuoteSuppressant-Prompt

Star

Prompt suppresses excessive use of scare quotes in LLM responses by framing use as failure to commit to precise language.

prompt-engineering llm-prompting precision-language clear-communication hedging-suppression scare-quotes language-calibration anti-rlhf high-fidelity-communication

Updated Jan 30, 2026

Add this topic to your repo

To associate your repository with the anti-rlhf topic, visit your repo's landing page and select "manage topics."

Learn more