opencode-configure

Configure OpenCode runtime settings in opencode.json and AGENTS.md.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines the process of configuring OpenCode, ensuring that your AI agent operates with the correct providers, models, permissions, and behavioral rules, preventing misconfigurations and security risks.

Core Features & Use Cases

  • Provider & Model Setup: Define and manage AI model providers and their default models.
  • Permission Management: Configure granular access controls for file system, shell commands, and tools.
  • Behavioral Rules: Establish project-specific instructions and constraints via AGENTS.md.
  • Use Case: When setting up a new project, use this Skill to define your preferred AI provider (e.g., OpenAI, Anthropic), set default models, restrict shell access to only necessary commands like npm test, and add project-specific workflow rules to AGENTS.md.

Quick Start

Configure your opencode.json to use the 'openai' provider with 'gpt-4o-2024-08-06' as the default model, referencing the API key from the OPENAI_API_KEY environment variable.

Frequently Asked Questions about opencode-configure

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

FAQPage Schema
How do I configure OpenCode provider and model settings?

Configure OpenCode provider and model settings by defining your preferred AI provider, such as OpenAI or Anthropic, and setting the default model in the `opencode.json` file. This ensures your AI agent operates with the correct runtime configuration and prevents misconfigurations.

What is the best way to restrict agent permissions for shell commands in OpenCode?

Restrict agent permissions for shell commands by configuring granular access controls in the `opencode.json` file. You can limit file system, tool, and shell access to only necessary commands like `npm test`, tightening security and preventing unauthorized actions.

Can I set project-specific behavioral rules for my AI agent?

Yes, you can set project-specific behavioral rules and workflow constraints by adding instructions to the `AGENTS.md` file. This establishes deterministic agent behavior guidelines tailored to your project's specific requirements and operating context.

Do I need to understand the `opencode.json` schema to manage runtime configurations?

Yes, understanding the `opencode.json` schema is required to define provider definitions, model defaults, and permission policies. This knowledge allows you to accurately troubleshoot OpenCode configuration issues and manage deterministic provider setup.

Why does my OpenCode configuration fail to apply the default AI model?

OpenCode configuration may fail to apply the default AI model if the `opencode.json` schema is misconfigured or the API key environment variable is missing. Troubleshoot by verifying provider definitions and ensuring the correct model selection syntax is used.