A Claude Code plugin that adds adversarial self-critique to every task. Claude scores its own work across multiple dimensions, identifies weaknesses, and proposes improvements — before you ship anything.

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

distilled Self-Critique refines the outputs of a LLM with only synthetic data

Autonomous deep-research platform powered by Hermes Agent. Multi-phase pipeline with parallel research, self-critique, iterative refinement, and real-time streaming.

MetaSC: Test-Time Safety Specification Optimization for Language Models

Reflexion agent with LangGraph — GPT-4.1 Nano self-critiques answers, generates Tavily search queries, revises with citations through up to 4 iterative improvement loops

"Merging Improves Self-Critique Against Jailbreak Attacks", code and models

ReAct agent with multi-hop planning, self-critique, and conversation memory over 150 arXiv ML papers — powered by BGE embeddings, FAISS, BM25, and Gemini 2.5 Flash.

Multi-agent AI pipeline that rewrites your resume for any job, scores it against ATS criteria, self-critiques the output, and generates a cover letter — powered by NVIDIA NIM (free tier). Streamlit + ChromaDB RAG.

Local-first benchmark for measuring reasoning improvement in small language models through critique and debate.

AI planning platform using LLM intrinsic self-critique.

Self-critic workflow for Claude Code to review its own output

Implementation of the Self-Critique Chain pattern for automated research paper summarization using Claude AI.

Two things: a working chain-of-thought + self-critique prompt (PROMPT.md), and a case study in how a leading question convinced a capable model to invent a confident framework for something that doesn't exist. The prompt is useful. The anatomy is cautionary.

Agentic CRM support system with automated data ingestion, structured RAG, and vector search