opencode-design-agents

Guide OpenCode agent creation and refinement through a question-and-answer process.

9|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/pantheon-org/tekhne --skill opencode-design-agents
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opencode-design-agents
Source: https://github.com/pantheon-org/tekhne/tree/main/skills/agentic-harness/opencode-toolkit/design-agents
Command: npx skills add https://github.com/pantheon-org/tekhne --skill opencode-design-agents

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the creation and improvement of OpenCode agents, ensuring they are well-defined, secure, and effective for their intended tasks.

Core Features & Use Cases

  • Agent Creation: Guides users through defining an agent's purpose, triggers, and persona.
  • Configuration Refinement: Helps set up agent modes, permissions, and tool access.
  • Troubleshooting: Assists in diagnosing and fixing issues with agent behavior and triggering.
  • Use Case: You need to create a new agent that can analyze code for security vulnerabilities. This Skill will guide you through defining its description, setting appropriate bash permissions, and writing its system prompt.

Quick Start

Use the opencode-design-agents skill to create a new agent for reviewing code.

Frequently Asked Questions about opencode-design-agents

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

FAQPage Schema
How do I create an OpenCode agent and configure its system prompt?

You can create an OpenCode agent by using a guided question-and-answer process to define the agent's scope, configure its mode, write the system prompt, and set up the AGENTS.md frontmatter.

What is the best way to restrict permissions for an AI agent?

To restrict AI agent permissions, you can define allowed-tools and set permission blocks during the agent configuration process, ensuring the agent only accesses delegated tools necessary for its specific tasks.

How do I configure subagent routing and Task tool delegation in OpenCode?

Configuring subagent routing and Task tool delegation involves setting the subagent type and defining tool delegation patterns within the agent file, allowing the primary agent to route tasks to specialized subagents.

Can I troubleshoot and improve an existing AI agent's triggering behavior?

Yes, you can troubleshoot and improve an existing AI agent by refining its description, adjusting its triggers, and updating the system prompt to ensure the agent activates correctly for its intended tasks.

Do I need to manually write AGENTS.md frontmatter to set up agent modes?

No, you do not need to manually write the frontmatter; the agent design process guides you through configuring agent modes and generating the AGENTS.md frontmatter automatically based on your answers.

Why is my OpenCode agent not triggering or delegating tasks correctly?

An OpenCode agent may fail to trigger or delegate tasks if its description is unclear, allowed-tools are misconfigured, or the subagent type and permission blocks are not properly set in the AGENTS.md file.