help

Lists vault capabilities with natural-language phrasings mapped to skills and CLI verbs.

2|1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/brandtam/rubber-ducky-legacy --skill help-brandtam
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: help
Source: https://github.com/brandtam/rubber-ducky-legacy/tree/main/src/skills/help
Command: npx skills add https://github.com/brandtam/rubber-ducky-legacy --skill help-brandtam

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Users of a rubber-ducky vault often forget which commands exist or how to phrase a request, and this Skill provides the canonical, on-demand index of every capability without burning tokens proactively. ## Core Features & Use Cases - Capability Discovery: Answers "what can you do?", "how do I ...?", and "commands" with a scannable, intent-grouped list of every skill and CLI verb. - Natural-Language Mapping: Maps user phrasings like "good morning", "I'm starting on ...", or "push this to <backend>" to the underlying skills such as /good-morning, /start, and /push. - Settings Reference: Documents every settings key (ingest., confirm.) with the exact rubber-ducky settings CLI commands and the risk-class confirmation model. - Use Case: A new user finishes onboarding and asks "what can you do?" — the skill renders the full grouped capability index, from daily routines to integrations, in one message. ## Quick Start Ask the assistant "what can you do?" or type "help" to get the full list of vault capabilities and the phrases that trigger each one.

Frequently Asked Questions about help

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I see what commands are available in rubber-ducky?▼

Type "help", "/help", or ask "what can you do?" and the help skill renders a grouped list of every capability. Each entry shows the natural-language phrasing first and the underlying skill or CLI verb second.

How do I change rubber-ducky settings like auto-ingest or confirmations?▼

Use the rubber-ducky settings CLI: get, set, enable, or disable followed by the setting path, such as ingest.auto_on_wrap_up or confirm.<transport>.<verb>. Every change is audit-logged to wiki/log.md.

What is the difference between soft and hard risk classes?▼

Soft actions are reads, drafts, and local vault writes that auto-execute. Hard actions send, post, or create on external systems and are gated by rubber-ducky confirm with a preview prompt unless overridden in settings.

Can I run rubber-ducky inside Obsidian instead of a terminal?▼

Yes. Install the Terminal community plugin in Obsidian, set its launch command to claude and the working directory to the vault root, then open a terminal pane. Behavior is identical to a standalone terminal.

Does the help skill show tips proactively during sessions?▼

No. Help is strictly invoked, never pushed — there are no proactive tips, daily banners, or inline reminders, so it costs zero tokens unless the user explicitly asks for it.