contrib-extract

What problem does it solve?

Papers often hide their real contributions beneath background text, inflated claims, and implementation details; this Skill extracts every distinct novelty and explains why it matters, where it appears, and what the key equation is. The output is structured to give a reader immediate clarity on "what is actually new" without wading through the full pipeline description.

Core Features & Use Cases

• Four-ingredient contribution blocks: For each contribution, produce a headline, type, motivation, an old-vs-new comparison when applicable, an intuition grounded in a scenario, and a core formula or mathematical object.
• Source-fidelity and equation cross-checking: Verifies formula symbols against page images and enforces a checklist-driven math explanation to avoid transcription errors.
• Use Cases: Rapid literature triage, preparing peer review summaries, generating contribution sections for a research reading group, or acting as the first stage in an automated paper-reader pipeline.

Quick Start

Ask the skill to list and explain every distinct contribution of the target paper, giving motivation, intuition with a scenario, an old-vs-new comparison if relevant, and the core formula for each contribution.