opencode

Delegate code implementation, refactoring, and review tasks to an external OpenCode CLI agent.

Updated Sep 1, 2021
One-click install
npx skills add https://github.com/unclehowell/unclehowell --skill opencode-unclehowell
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opencode
Source: https://github.com/unclehowell/unclehowell/tree/main/skills/autonomous-ai-agents/opencode
Command: npx skills add https://github.com/unclehowell/unclehowell --skill opencode-unclehowell

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Offload implementation, refactoring, and review work to an external autonomous coding agent so engineers can parallelize tasks and manage long-running, iterative code changes without manual terminal supervision.

Core Features & Use Cases

  • One-shot automation: Run bounded commands to perform targeted changes or analysis and exit with machine-readable output.
  • Interactive long-running sessions: Start TUI-backed background sessions with session IDs, submit follow-ups, poll logs, and manage progress.
  • PR review & isolation: Review pull requests in temporary clones or separate worktrees to avoid collisions and perform safe commits.
  • Use Case: Run an isolated OpenCode session to refactor an authentication module, add tests, and produce a summary of changed files and remaining issues.

Quick Start

Use the opencode skill to run a one-shot command that adds retry logic to the repository's API client, updates tests, and summarizes the files changed and test results.

Frequently Asked Questions about opencode

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

FAQPage Schema
How do I delegate code refactoring to an autonomous agent without staying in the terminal?

You can delegate code refactoring to an autonomous agent by offloading implementation work to the external OpenCode CLI. This allows engineers to parallelize tasks and manage iterative code changes without manual terminal supervision.

Can I review pull requests in isolated worktrees to prevent file collisions?

Yes, you can review pull requests in isolated worktrees to prevent collisions. The CLI reviews PRs in temporary clones or separate worktrees, enabling safe commits and avoiding branch conflicts.

How do I manage long-running interactive coding sessions for iterative feature implementation?

To manage long-running interactive sessions, start TUI-backed background sessions with unique session IDs. You can submit follow-up prompts, poll logs, and track progress without keeping the terminal open.

Do I need the OpenCode CLI installed to automate bounded code changes?

Yes, you need the OpenCode CLI installed and authenticated to automate bounded code changes. The environment must also have pty enabled to support interactive TUI sessions for autonomous execution.

What's the best way to run an isolated session to add tests and summarize changed files?

The best way to add tests and summarize changed files is running a one-shot command through the CLI. It performs targeted changes, executes tests, and exits with machine-readable output summarizing the results.

Does the autonomous CLI agent support file attachments and explicit model selection?

Yes, the autonomous CLI agent supports file attachments and explicit model selection. You can attach necessary files directly to the agent and specify the exact model to use for the delegated coding task.