TL;DR
A plugin that supports the entire academic research process (planning-writing-review-revision-finalization) within the Claude Code environment. It is a human-AI collaboration tool that prevents major failure modes of AI research, such as citation verification and logical consistency checks.
Key features
Full pipeline: Step-by-step guidance from research planning (Socratic dialogue), literature search, draft writing, review, revision, to final completion
Citation verification: Provides trust-chain, location-based anchors, and claim audit features to address the hallucinated citation problem highlighted in a large-scale study by Zhao et al. (2026)
Integrity gates: A blocking checklist with 7 modes to prevent AI research failure modes such as implementation bugs, hallucinated results, and methodology manipulation
Style correction: Learns style from the user's past research to reduce machine-generated feel and correct to a natural writing style
Quality check: Writing Quality Check function that detects and improves machine-generated sentence patterns
When to use it
When you are writing an academic paper and using Claude Code
When you want to increase productivity while preventing AI citation hallucinations or logical errors
When you need a systematic workflow from research planning to final submission
When you want to improve the quality of machine-generated text and maintain a natural academic writing style