Personal terminal and Claude Code configuration.
terminal/
.tmux.conf # tmux config (mouse, scrollback, sync-panes)
.blerc # ble.sh config (suppress EOF, disable bell)
starship.toml # Starship prompt theme
claude/
settings.json # Claude Code settings (plugins, hooks, env vars)
hooks/
notify.sh # Desktop + sound notifications on Stop/Permission/Notification
skills/
cs-research-team/ # Coordinate CS PhD research with agent teams
lit-review-team/ # Run literature reviews with agent teams
lit-review-team-old/ # Previous lit-review skill version
git clone https://github.com/Asonjay/claude-skills.git dotfiles
cd dotfiles
./install.shThe interactive installer lets you pick what to set up:
- Terminal — installs tmux, ble.sh, starship if missing; symlinks configs; optionally adds shell init to
.bashrc - Claude Code — copies settings/hooks, symlinks skills, adds the
anthropics/claude-codemarketplace, checks for notification dependencies (paplay,notify-send) - Everything
Existing files are backed up before being replaced.
These are auto-installed by Claude Code when enabled in settings.json:
| Plugin | Marketplace | What it provides |
|---|---|---|
superpowers |
claude-plugins-official |
Development workflows: brainstorming, TDD, debugging, code review, git worktrees, parallel agents |
document-skills |
anthropic-agent-skills |
PDF, DOCX, PPTX, XLSX manipulation, web artifacts, MCP builder |
frontend-design |
anthropics-claude-code |
Production-grade frontend UI with bold design choices, typography, animations |
planning-with-files |
planning-with-files |
Manus-style persistent markdown planning, progress tracking, knowledge storage |
Plugin format: <plugin-name>@<marketplace-name>. Claude Code resolves marketplace names to GitHub repos and downloads them to ~/.claude/plugins/cache/.
The frontend-design and planning-with-files plugins require extra marketplaces (the installer adds them automatically, or run claude plugin marketplace add anthropics/claude-code and claude plugin marketplace add OthmanAdi/planning-with-files).
Spawns literature, implementation, experiments, and writing agents:
/cs-research-team ./plans/my-research.md
/cs-research-team ./plans/my-research.md 4 # specify team sizeSpawns reader, scaffolder, and synthesizer agents:
/lit-review-team ./plans/review-plan.md
/lit-review-team ./plans/review-plan.md 3 # specify team sizeAgent teams are experimental. The settings.json already includes:
{
"env": {
"CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1"
}
}Or export in your shell:
export CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1