help-me

Inventories opencode agents, skills, and commands and recommends workflows for common tasks.

Updated Aug 2, 2026
One-click install
npx skills add https://github.com/jakestrouse00/opencode-toolkit --skill help-me-jakestrouse00
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: help-me
Source: https://github.com/jakestrouse00/opencode-toolkit/tree/main/skills/help-me
Command: npx skills add https://github.com/jakestrouse00/opencode-toolkit --skill help-me-jakestrouse00

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a project accumulates many opencode agents, skills, and commands, it becomes hard to know which tool fits a given task. This Skill reads the live definitions under .opencode/ and gives a guided tour of what each tool does, when to use it, and how they chain together. ## Core Features & Use Cases - Live inventory: Reads .opencode/agent/*.md, .opencode/skills/*/SKILL.md, and .opencode/commands/*.md at run time, summarizing each tool's job, triggers, permissions, and hand-offs. - Workflow recommendations: Maps common requests (planning a change, committing code, triaging plans, adding agents or tools) to concrete step-by-step tool sequences. - Capability matrix: Explains the agent vs skill vs command layering so users pick the right layer for read-only research, guided edits, or fixed repeatable actions. - Use Case: A new contributor asks "what can you do?" and receives a concise tour of the toolkit plus a recommended workflow for their specific goal, without any files being modified. ## Quick Start Ask the assistant "help me pick the right agent or skill for planning a multi-file change" to get a live inventory and a recommended workflow.

Frequently Asked Questions about help-me

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

FAQPage Schema
How do I find out which opencode agents and skills are available in my project?

Invoke the help-me skill, which reads every agent definition under .opencode/agent/*.md and every SKILL.md under .opencode/skills/*/ at run time. It summarizes each tool's job, trigger phrases, permissions, and hand-offs based on the live files on disk.

What is the difference between an opencode agent, skill, and command?

Agents run as sub-agents with their own context and permission allowlists, skills run in the main session as step-by-step workflows, and commands are user-invoked TUI shortcuts typed by name like /init. The skill's capability matrix maps common jobs to the correct layer.

Does the help-me skill modify any files in my repository?

No, the skill is strictly read-only. It never edits files, runs commands, or commits anything; it only reads AGENTS.md and the .opencode/ definitions to produce a summary and recommendations.

What happens if the toolkit inventory in the skill is out of date?

The skill re-reads the agent, skill, and command files live at run time and trusts the files on disk over its cached summaries. If tools were added, removed, or renamed, it reports what is actually present.

When should I use an agent instead of a skill in opencode?

Use an agent for read-only bulk research or autonomous analysis that returns a report, and a skill for guided edits you want to watch, such as planning, wiring, or committing. Commands fit fixed repeatable actions invoked by name.