메뉴
Use when the user asks to run OpenCode CLI in non-interactive mode (opencode run) or references OpenCode for AI-assisted coding tasks.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | opencode |
| description | Use when the user asks to run OpenCode CLI in non-interactive mode (opencode run) or references OpenCode for AI-assisted coding tasks. |
Before running OpenCode, follow the shared logging and summary protocol in ../_shared/agent-cli.md.
command -v opencodeplan for read-only analysis. For implementation/editing tasks, first use superpowers:using-git-worktrees to create a fresh git worktree, then run OpenCode there with --agent build when approved.the current host's user-question or approval mechanism before using the build agent for write operations.-m, --model <provider/model> - Model to use-c, --continue - Continue the last session-s, --session <id> - Session ID to continue--agent <agent> - Agent to use (build for full access, plan for read-only)-f, --file <files...> - File(s) to attachfull.md, and require the prompt to write summary.md.| Use case | Command pattern |
|---|---|
| Full access in fresh worktree only (write/edit/bash) | opencode run --agent build "prompt" |
| Read-only analysis | opencode run --agent plan "prompt" |
| Continue last session | opencode run --continue "follow-up" |
| Continue specific session | opencode run --session <id> "prompt" |
| Use specific model | opencode run --model anthropic/claude-sonnet-4-20250514 "prompt" |
| Attach files | opencode run -f file.py "prompt" |
| List models | opencode models |
| List sessions | opencode session list |
| List available agents | opencode agent list |
| Agent | Permissions | Use Case |
|---|---|---|
build | Full access: write, edit, bash | Active development in a fresh worktree |
plan | Read-only: no write/edit, bash asks first | Code analysis, planning, exploration |
# Read-only analysis with plan agent
opencode run --agent plan "Analyze the codebase architecture"
# Full access with build agent in a fresh worktree only
opencode run --agent build "Fix the bug in main.py"
# Continue last session
opencode run --continue "What else can you improve?"
# Use specific model
opencode run --model anthropic/claude-sonnet-4-20250514 "Refactor this code"
# Capture stdout/stderr according to ../_shared/agent-cli.md
opencode run command, immediately use the current host's user-question or approval mechanism to confirm next steps, collect clarifications, or decide whether to resume with --continue.opencode --version or an opencode command exits non-zero; request direction before retrying.opencode debug for troubleshooting if needed.Use when installing, reinstalling, auditing, or preparing Brain-managed Superpowers, Spec Kit, and OpenSpec workflow arsenals under a project-local `.brain/vendor` directory so Brain can inspect and route them without exposing their raw Codex skills/prompts at the project root.
Use when the user asks to check, add, or manage macOS Calendar events.
Use when the user asks to send an email via the macOS Mail app.
Use when the user asks to send a text message or iMessage via the macOS Messages app.
Use when a task involves web search, current information, source verification, citations, or entity/person/paper/project disambiguation.
Use when the task involves `exec_command` escalation, `prefix_rule`, sandbox permissions, writable roots, or Codex platform-specific operational patterns.