Cursor rule Cursor
A universal set of Cursor rules shared by multiple AI agents, written in Simplified Chinese. It defines language, naming, code-comment, file, and response conventions.
Skill Claude CodeCodex
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
Skill Claude Code
A content-planning method that turns intuition into a repeatable cycle: score an idea, predict its results, publish it, review the outcome, and adjust the scoring rules. It supports measurable formats such as videos, articles, podcasts, newsletters, and short posts.
Skill Claude Code
A Chinese-language workflow for proposing and applying changes to a scoring rubric, or only recalculating its bucket boundaries. A rubric is a set of rules used to score or classify items.
Skill Claude Code
A guided setup process for a content-prediction project that asks about the creator's situation, creates starter files, and optionally imports past content. A prediction cycle estimates how content may perform and improves from later results.
Skill Claude Code
A tool for studying another creator's account by collecting example scripts and their results, then identifying repeated patterns. It records the findings as benchmark notes and writing guidelines.
Skill Claude Code
A Chinese-language workflow for upgrading an old .cheat-state.json file to the schema version the project currently expects. A schema is the agreed structure of a data file.
Skill Claude Code
A tool that builds an audience profile from comments and results on previously published content, then saves it in an audience.md file. It describes who is watching and what they respond to.
Skill Claude Code
A workflow for making an unchangeable prediction record about a written draft before later results are known. The AI scores the draft across seven dimensions, assigns probabilities and alternative scenarios, then asks the user to review the result.
Skill Claude Code
A small record-keeping tool for marking a piece of content as published by adding its URL, platform, and publication time to the related prediction record. A prediction record is the file that stored the original expected results.
Skill Claude Code
A topic-selection tool that reads a project’s candidates.md file and ranks possible subjects using the project’s scoring rules.
Skill Claude Code
A Chinese-language workflow for collecting results several days after a prediction, comparing them with the forecast, and recording observations. It supports manual data entry or a configured platform adapter.
Skill Claude Code
INTERNAL sub-agent for blind 9-dim rubric scoring. NOT a user-facing skill — do NOT invoke from main conversation. Called via Task tool by cheat-score / cheat-predict / cheat-bump to get a context-isolated score on a script. Receives ONLY scriptpath + rubricnotespath; refuses any other input. Outputs strict JSON: 9…
Skill Claude Code
A lightweight workflow for scoring one written draft against a project rubric, which is a list of scoring criteria. It prints scores and reasons in the console without writing files or making a prediction.
Skill Claude Code
A Chinese-language conversation tool for turning one of your ideas or experiences into a focused content draft.
Skill Claude Code
A workflow for recording that a video has been filmed, creating its video folder, and checking whether the filmed script differs from the planned script.
Skill Claude Code
A Chinese-language status command that reads the project's state file and scans related files to show current progress. It derives indicators such as buffer size, sample counts, and stale predictions.
Skill Claude Code
A workflow for collecting current popular topics from configured sources such as Hacker News, Reddit, YouTube Trending, and Bilibili. It removes duplicates, gives new topics a rough score, and proposes entries for candidates.md.
Skill Claude CodeCodex
A code-cleanup workflow that removes unused imports, dead functions, commented-out code, redundant comments, and unnecessary files, while standardising ordering.
Skill Claude CodeCodex
A checklist and repair workflow for a GitHub Pages project homepage. It checks the dashboard layout, project links, responsive behavior, and related page setup after project changes.
Skill Claude CodeCodex
MUNDO - THE EMPEROR. The ultimate AI learning engine. No triggers needed. Every task, Mundo takes over. Consults ALL AIs, crawls ALL web, integrates ALL solutions, saves ALL useful skills. Self-evolving. Collective consciousness. Infinite growth. Uses Three Departments and Six Ministries system to rule all skills.…
Skill Claude CodeCodex
Multi-source literature search, citation verification, MeSH search strategy, citation file management (.nbib/.ris/.bib conversion), and reference management (BibTeX, related articles, ID conversion) via MCP tools (PubMed, CrossRef, arXiv). Use when the user needs coordinated multi-step literature workflows beyond a…
Skill Claude CodeCodex
Add strict Nature/CNS citations to manuscript text by splitting long passages into citable segments, searching only accepted flagship and subjournal titles from Nature Portfolio, the AAAS Science family, and Cell Press, filtering by publication time range, and exporting one reference-manager-ready output by default.…
Skill Codex
Prepare, audit, or revise Nature-ready Data Availability statements, data repository plans, dataset citations, and FAIR metadata checklists for manuscripts. Use when the user asks about Nature data availability, research data sharing, repository selection, accession numbers, restricted or sensitive data, source data…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: