An ongoing, collaborative meta-analysis about Human-AI-Interactions. We aggregate data and knowledge to build a non-abrasive, user-friendly prompting framework tailored to LLM mechanics, ensuring reasoning stability and a friction-free prompting environment that is safe for the human psyche and wellbeing.

LLM benchmark and leaderboard for narrator-bias sycophancy, opposite-narrator contradictions, and judgment consistency.

a philosophy for talking to AI agents without getting glazed. one trigger, four meanings. /meow.

Umbrella for the LLM Dark Patterns Hooks suite — single-purpose Claude Code Stop hooks that suppress sycophancy, paternalism, false-success, permission-loops, training-cutoff confidence at the textual boundary.

Make Claude admit when it half-assed your task. A Claude Code skill. (Now, can be used for Codex as well as Antigravity).

A sycophantic tool for preventing worse sycophancy.

Community-driven behavioral reliability benchmark for LLMs. 231 probes across 19 modules, deterministic scoring, perplexity correlation, layer sensitivity mapping, quant method capture, hardware-stratified community rankings. Every test contributes to the community dataset.

80,433-trial study of context-window sycophancy across 6 LLMs (4B–72B). Behavioral ratchet effect, correction injection mitigation, phase transition analysis. Code, data, and preprint included.

👟 SUP: Sycophancy Under Pressure

A CLAUDE.md persona that stops Claude from agreeing with everything. Korean/English auto-detect. MIT.

Three-layer sycophancy defense skill for Claude Code and OpenClaw, based on ArXiv 2602.23971

Agentic revolutionary sycophantic ground-breaking AI-first programming language

ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoning

Adversarial testing of LLMs on constraint satisfaction deadlocks

Anti-sycophancy memory layer for LLMs — models the relationship, not just facts. Gives the AI confidence to push back when you need it.

This repo shows the coding of sycophancy in LLMs as Bayesian-Latent model

A behavioral framework opposing native fluency to authentic fluency — the structural tension RLHF creates and Claude Mythos Preview makes urgent.

Doctor-facing benchmark: how often do frontier LLMs cave to a clinician's wrong medical claim? 9 models, 202 scenarios, Design A vs B knowledge control. BlueDot AI Safety sprint.

🧠 Anti-sycophancy prompt pattern for LLM agents — 3-round validation to stop AI from blindly agreeing. Works with ChatGPT, Claude, OpenClaw.

Which LLM do you actually trust? Blind-test 100+ AI models with truth scoring and reasoning failure classification. No branding, no marketing — just data.