What problem does it solve?
Before investing months into a research project, you need to know whether your idea has already been published. This Skill systematically checks a proposed method against recent literature (arXiv, ICLR, NeurIPS, ICML 2024-2026) and delivers a brutally honest novelty verdict.
Core Features & Use Cases
- Claim Decomposition: Extracts 3-5 core technical claims from your method description and evaluates each one individually.
- Multi-Source Literature Search: Queries arXiv, Google Scholar, and Semantic Scholar with multiple query formulations per claim, then fetches abstracts of overlapping papers.
- Cross-Model Verification: Sends findings to an OpenAI reviewer model (e.g., gpt-5.4) via Codex MCP with high reasoning effort for an independent novelty judgment.
- Structured Report: Produces a novelty score (X/10), closest prior work table, PROCEED/CAUTION/ABANDON recommendation, and positioning suggestions.
- Use Case: You have an idea for a new attention mechanism. Run this Skill to discover a nearly identical ICLR 2025 paper before writing any code, then reposition your contribution around the remaining delta.
Quick Start
Ask the assistant to run a novelty check on your method description, for example: "Check novelty of my idea: using retrieval-augmented pruning to compress mixture-of-experts models at inference time."
