-
Notifications
You must be signed in to change notification settings - Fork 2
CLI
github-actions[bot] edited this page Apr 15, 2026
·
6 revisions
All dots-* CLI helpers with subcommands. For full details, see docs/CLI_HELPERS.md.
| Command | Purpose |
|---|---|
dots-doctor |
Validate baseline compliance and required tooling |
dots-skills |
Manage AI skills across all installed AI tools |
dots-devcompanion |
Dev companion background job queue and runner |
dots-llm-server |
Manage local LLM server profiles |
dots-bootstrap |
Guided bootstrap output with optional --apply
|
dots-sync-ai |
Verify shared AI directories are present |
dots-update-check |
Check whether local baseline is outdated |
dots-loadenv |
Show/emit opt-in env loading for ~/.config/dots-ai/env.d/*.env
|
dots-skills list List installed skills and status per AI tool
dots-skills sync Regenerate symlinks for all skills
dots-skills install <name> Install a skill from the registry
dots-skills check Validate required CLI tools per skill
dots-skills add npm:<pkg> Add an npm skill to the registry
dots-devcompanion enqueue <id> Queue a background job
dots-devcompanion run-once Process the next queued job
dots-devcompanion status Show queue status
dots-devcompanion done <id> Mark a job as complete
Note
Requires an NVIDIA GPU. Profiles share GPU memory — only one runs at a time.
dots-llm-server install Pull images and download models
dots-llm-server start [profile] Start coding (default) or reasoning
dots-llm-server stop Stop all LLM services
dots-llm-server switch [profile] Toggle or switch profile
dots-llm-server status Show service status
dots-llm-server logs [profile] Follow logs
| Profile | Engine | Port | Model | Speed |
|---|---|---|---|---|
coding |
vLLM (Docker) | 8000 | Qwen2.5-Coder-7B-AWQ | ~400 tok/s |
reasoning |
Ollama | 11434 | DeepSeek-R1-7B-Q4 | ~26 tok/s |
dots-update-check # check for updates
chezmoi apply --dry-run # preview changes
chezmoi apply # apply
dots-doctor # validate
dots-skills check # verify skill dependencies- Technical Quickstart — full setup guide
- Skills System — skill management details