Automates LaTeX paper drafting from outline to submission-ready PDF, one section at a time. Works with outputs from paper-plan and paper-figure skills, pulling in your research narrative and generated figures. Ships with templates for major ML venues (ICLR, NeurIPS, ICML) and IEEE conferences, handling anonymous submissions and page limits automatically. The standout feature is real BibTeX fetching from DBLP and CrossRef, which eliminates hallucinated citations without any setup. Optional style-ref mode lets you point at an exemplar paper and match its structural rhythm (section count, theorem density, citation style) without copying content. Includes cross-model review via GPT for each section and explicit gap markers when your evidence doesn't cover what the outline implies you need.
$ npx -y skills add wanshuiyin/auto-claude-code-research-in-sleep --skill paper-write --agent claude-codeInstalls into .claude/skills of the current project.
Before each reviewer slot, follow Selectable Reviewers: resolve invocation →
.aris/reviewer.json→ Claude Code default (Codex MCP) withreviewer_route.pythrough the canonical helper resolver, preserving this slot's tier, fresh contexts, source/image inputs and deterministic gates. Codex examples/pins below apply only tobackend=codex; other routes use their own first-call/reply contract. Save the route and actual returned identity; unavailable selections areREVIEW_UNAVAILABLE. Classify acceptance from actual families/completion/scope. Forward invocation reviewer/model/effort overrides to reviewer-bearing children. Writer selection is separate; follow the writing-worker guidance below when drafting or revising prose.
Draft a LaTeX paper based on: $ARGUMENTS
Sponsors