codex-plugins

Spawn a remote-control Claude Code instance from inside the current Codex session via detached tmux, and return its /remote-control URL. Use when the user wants to fire off a sibling Claude Code session they can drive in parallel from a browser (e.g. delegate a long task to CC while keeping the current Codex session free). Only works on the local host. Codex itself does not have a /remote-control equivalent, so this is a cross-tool helper (Codex → Claude Code), not a Codex-to-Codex spawner.

Manage TODO.md in the current project — show, add, complete, delete, update items via natural language. Use whenever the user mentions a TODO action ("加一条 X" / "X 完成了" / "删掉 X" / "X 改成 Y" / "add a todo" / "mark done" / "remove" / "check my todos"), references TODO.md, or asks to see today's calendar. Also handles auto-archive of completed items older than 7 days to `~/.claude/memory/<完成日期>.md`.

Academic-paper editing for Overleaf-synced LaTeX projects (ML / CS / arxiv submissions). Invoke when the user wants to pull / edit / review / polish a paper section (abstract, intro, method, related work, experiments, conclusion, limitations); apply red-mark feedback wrapped in `\textcolor{red}{...}`; do bilingual zh/en side-by-side review; audit numerical claims against the codebase; or check writing-style compliance. Enforces NeurIPS / ICML / ICLR / COLM venue conventions and the user-decision boundaries below (naming, title, structural forks, paragraph deletion). Does NOT handle — blog posts, README, slides, response letters, rebuttals.

Matplotlib templates for paper / blog / report figures with a Google-brand palette and Palatino body font (matches arxiv mathpazo). Use when the user asks for a chart they will save and paste into a paper, slide deck, or write-up — bar, boxplot, line, scatter variants. Templates ship pre-genericized (Model A/B, Metric A/B); replace with real names when applying. Skip for one-off exploratory plots inside notebooks where styling does not matter.

Pre-submission self-review of your own AI/ML paper Overleaf draft. Takes a project name (resolves to ~/Documents/docs/project/<name>/), recursively ingests the LaTeX source, and produces a reviewer-style critique markdown report simulating a top-tier ML venue reviewer (NeurIPS / ICML / ICLR / CVPR). Trigger on natural-language "review my paper draft", "self-review before submission", "give reviewer feedback on my paper". Do NOT trigger for peer-reviewing others' papers, code review, PR review, or paragraph-by-paragraph editing iteration (that's a separate workflow).

Default writing standards. Use whenever producing English prose the user will send or publish — emails, message drafts, blog posts, tweets, documentation, grant proposals, and conference/journal paper drafts (NeurIPS, ICML, ICLR, ACL, CVPR, COLM, EMNLP, arxiv). Applies 17 canonical rules everywhere; 18 page-capped additions (RULE-P1..P18) for page-constrained papers; 2 audit-time rules (RULE-A1..A2) only when reviewing a compiled PDF for submission. Treat canonical and page-capped rules as peer constraints applied at generation time; audit rules trigger at final-pass review, not during drafting.