Agent Skill
2/7/2026

xlsx

Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

B
bahayonghang
7GitHub Stars
1Views
npx skills add bahayonghang/my-claude-code-settings

SKILL.md

Namexlsx
DescriptionUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

My Claude Code Settings

License: MIT

Cross-platform AI content repository with a Rust management workspace.

This repository is organized around two top-level areas:

  • content/: installable skills, commands, agent definitions, runtime files, and external-skill registry data
  • mcs/: the Rust workspace that powers discovery, install, diff, sync, TUI, and Web workflows

Quick start

git clone https://github.com/bahayonghang/my-claude-code-settings.git
cd my-claude-code-settings

# Terminal UI
just mcs

# Web UI + backend
just web

# Documentation site
just doc

Repository layout

.
├── content/
│   ├── skills/            # First-party skill catalog + external-skills/ registry fragments
│   ├── commands/          # Slash commands and workflow sources
│   ├── agents/            # Markdown agent definitions
│   ├── hooks/             # Runtime hook assets
│   └── memorys/           # Platform-specific runtime memory / prompt files
├── docs/                  # VitePress documentation site
├── mcs/                   # Rust workspace: mcs-core, mcs-tui, mcs-web
├── platforms.toml         # Platform install mapping
└── justfile               # Common entrypoints

The curated third-party registry now lives at content/skills/external-skills/.

Skill categories

The first-party catalog under content/skills/ currently uses these categories:

  • academic-skills
  • ai-llm-skills
  • diagram-skills
  • document-skills
  • git-github-skills
  • media-skills
  • skill-meta-skills
  • tech-stack-skills
  • workflow-skills

For the browsable catalog, use the docs site or MCS.

MCS workspace

The Rust workspace in mcs/ currently contains:

  • mcs-core: shared discovery, metadata, install, path, migration, and prompt logic
  • mcs-tui: ratatui/crossterm terminal UI
  • mcs-web: Axum backend plus React frontend hosting

Useful commands:

just mcs
just mcs-dev
just mcs-web
just mcs-web-server
just mcs-web-test

Platform model

Platform paths are defined by:

  1. built-in defaults in mcs-core
  2. platforms.toml in this repository
  3. optional user overrides in ~/.config/myclaude/platforms.toml

Not every platform has its own independent source tree. For example, commands may install from a fallback source directory even if the target platform installs them to a different folder.

Documentation

The VitePress site in docs/ covers:

  • installation
  • MCS TUI
  • MCS Web
  • MCS architecture
  • commands
  • runtime files
  • external skills
  • skill catalog pages in English and Chinese

For the Codex CLI skill specifically, see docs/skills/ai-llm-skills/codex.md and docs/zh/skills/ai-llm-skills/codex.md. Those pages track the current Codex CLI syntax and now document the gpt-5.4 default model plus the recommended codex exec / live web-search patterns.

Direct skill-only install

To install all skills non-interactively to specific agents (Universal, Antigravity, Claude Code, iFlow CLI, Kiro CLI, Qwen Code, Trae, Trae CN):

npx skills add bahayonghang/my-claude-code-settings/content/skills --skill '*' -g -y -a universal -a antigravity -a claude-code -a iflow-cli -a kiro-cli -a qwen-code -a trae -a trae-cn

If you only want the first-party skill catalog and do not need the full repository workflow:

npx skills add bahayonghang/my-claude-code-settings/content/skills

License

MIT

Skills Info
Original Name:xlsxAuthor:bahayonghang