compare-literature

What problem does it solve?

Compare task results against published literature and write the comparison output to a standardized report file, enabling reproducible validation of AI research findings.

Core Features & Use Cases

• Context gathering: ingest task objectives, task.json, and results files to prepare a literature-based benchmark.
• Published results synthesis: locate and extract relevant numbers from research files and papers to build a comparable baseline.
• Report construction: generate results/compare_literature.md with a structured, citation-backed table and narrative.
• Use Case: researchers validating an autonomous task by contrasting its metrics with published baselines to quantify improvement or gaps.

Quick Start

Run the compare-literature skill on a task ID to generate the comparison report document.