What problem does it solve?
Turning scattered research artifacts—narrative reports, auto-review conclusions, and experiment results—into a coherent, venue-compliant paper outline is time-consuming and error-prone. This Skill automates the planning phase by extracting claims, mapping them to evidence, and producing a page-budgeted outline ready for drafting.
Core Features & Use Cases
- Claims-Evidence Matrix: Extracts core claims from NARRATIVE_REPORT.md, GPT54_AUTO_REVIEW.md, and experiment JSON files, then maps every claim to supporting evidence.
- Venue-Aware Structure: Selects paper type (empirical, theory, method) and enforces page budgets for ICLR, NeurIPS, ICML, CVPR, ACL, AAAI, ACM, and IEEE venues, including IEEE's rule that references count toward the limit.
- Figure and Citation Planning: Produces a detailed figure plan (including a hero figure specification) and a verified citation plan that flags unverified references.
- Automated Review Loop: Sends the outline to GPT-5.4 via Codex MCP for scored feedback on logic, claim-evidence alignment, and page feasibility before finalizing PAPER_PLAN.md.
- Use Case: After finishing experiments for an ICLR submission, run the Skill to convert your review conclusions and result JSONs into a 9-page section-by-section outline with figure and citation plans.
Quick Start
Ask the assistant to generate a paper outline for an ICLR submission from the narrative report and experiment results in the current project directory.
