Opencode

OpenCode (open-source terminal coding agent by SST). Reads AGENTS.md natively. Rulebook generates opencode.json plus .opencode/commands, .opencode/agents, .opencode/skills.

11 stars

byhivellm

View on GitHub Installation ↓

Best use case

Opencode is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

OpenCode (open-source terminal coding agent by SST). Reads AGENTS.md natively. Rulebook generates opencode.json plus .opencode/commands, .opencode/agents, .opencode/skills.

Teams using Opencode should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/opencode/SKILL.md --create-dirs "https://raw.githubusercontent.com/hivellm/rulebook/main/templates/skills/cli/opencode/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/opencode/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How Opencode Compares

Feature / Agent	Opencode	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

OpenCode (open-source terminal coding agent by SST). Reads AGENTS.md natively. Rulebook generates opencode.json plus .opencode/commands, .opencode/agents, .opencode/skills.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

<!-- OPENCODE:START -->
# OpenCode Rules

**Tool**: [OpenCode](https://opencode.ai) — open-source terminal coding agent.

## Generated surface area

When Rulebook detects OpenCode (`opencode.json` / `.opencode/` / `opencode`
binary on PATH), it generates:

- `opencode.json` with `$schema`, `mcp.rulebook`, and `instructions`.
  Existing user keys (model, theme, permission, custom agents) are preserved.
- `.opencode/commands/<name>.md` for every user-invocable Rulebook slash
  command, with OpenCode-shaped frontmatter (`description`).
- `.opencode/agents/<role>.md` for every Rulebook role agent
  (researcher, implementer, tester, code-reviewer, architect, docs-writer,
  security-reviewer, build-engineer, team-lead, etc.) with model tier
  mapped (`haiku → claude-haiku-4-5`, `sonnet → claude-sonnet-4-6`,
  `opus → claude-opus-4-7`) and a synthesized `permission` block.
- `.opencode/skills/<normalized-name>/SKILL.md` for every Rulebook dev
  skill, name-normalized to `[a-z0-9](-[a-z0-9])*` (≤64 chars) and
  description bounded to ≤1024 chars.
- `.opencode/.rulebook-managed.json` — sidecar listing managed keys so
  `rulebook update` knows what to refresh vs. preserve.

## Idempotency

Files carrying the `<!-- RULEBOOK:START -->` marker are managed and
refreshed on every `rulebook update`. Files without the marker are
treated as user-owned and left untouched.

## Sequential editing

OpenCode follows the same Rulebook discipline as Claude Code:

- Edit files one at a time.
- For 3+ files, decompose into 1–2 file sub-tasks.
- Diagnostic-first: type-check / lint before tests.

## Quality gates

- Type-check / lint must be clean.
- Tests must pass.
- Coverage ≥95%.
- Pre-commit hooks enforce these — never `--no-verify`.

## MCP

`mcp.rulebook` in `opencode.json` exposes `rulebook_task_*`,
`rulebook_memory_*`, `rulebook_decision_*`, `rulebook_knowledge_*`,
`rulebook_learn_*`. Prefer MCP tools over shell commands when both
exist.

## Workflow

1. Reference `AGENTS.md` and `.rulebook/specs/*.md`.
2. Use `/rulebook-task-create`, `/rulebook-task-list`, `/handoff`.
3. Delegate to role agents (`@researcher`, `@implementer`, `@tester`).
4. Run quality checks before committing.
5. Save learnings (`rulebook_memory_save`, `rulebook_knowledge_add`).

## Documentation

- [opencode.ai/docs/rules](https://opencode.ai/docs/rules/)
- [opencode.ai/docs/config](https://opencode.ai/docs/config/)
- [opencode.ai/docs/mcp-servers](https://opencode.ai/docs/mcp-servers/)
- [opencode.ai/docs/commands](https://opencode.ai/docs/commands/)
- [opencode.ai/docs/agents](https://opencode.ai/docs/agents/)
- [opencode.ai/docs/skills](https://opencode.ai/docs/skills/)
- [opencode.ai/docs/permissions](https://opencode.ai/docs/permissions/)

<!-- OPENCODE:END -->

Related Skills

We are still matching the closest adjacent skills for this page. In the meantime, continue through the full directory.