What problem does it solve? Turning a vague research direction into both a focused paper proposal and a concrete experiment roadmap normally requires running separate workflows and manually keeping them consistent. This Skill chains method refinement and experiment planning into a single pipeline so the final proposal, review history, and validation plan stay aligned. ## Core Features & Use Cases - End-to-End Pipeline: Runs the research-refine stage to stabilize the method thesis, then the experiment-plan stage to build a claim-driven validation roadmap. - Planning Gate: Enforces a checkpoint that confirms the final thesis, dominant contribution, and rejected complexity before any experiments are planned. - Structured Deliverables: Produces FINAL_PROPOSAL.md, REVIEW_SUMMARY.md, REFINEMENT_REPORT.md, EXPERIMENT_PLAN.md, EXPERIMENT_TRACKER.md, and PIPELINE_SUMMARY.md under refine-logs/. - Use Case: A researcher with a rough idea for a video compression paper asks to do it end-to-end ("串起来"), and receives a refined one-contribution proposal plus an ordered experiment plan with budgets, decision gates, and must-run ablations. ## Quick Start Run the research-refine-pipeline on my research direction about machine-oriented video tokenization to produce both the final proposal and the experiment plan.
